diff options
| author | Felix Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> | 2023-09-27 17:46:17 +0100 |
|---|---|---|
| committer | felixjohnny.thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> | 2023-09-28 12:08:05 +0000 |
| commit | afd38f0c617d6f89b2b4532c6c44f116617e2b6f (patch) | |
| tree | 03bc7d5a762099989b16a656fa8d397b490ed70e /arm_compute | |
| parent | bdcb4c148ee2fdeaaddf4cf1e57bbb0de02bb894 (diff) | |
| download | ComputeLibrary-afd38f0c617d6f89b2b4532c6c44f116617e2b6f.tar.gz | |
Apply clang-format on repository
Code is formatted as per a revised clang format configuration
file(not part of this delivery). Version 14.0.6 is used.
Exclusion List:
- files with .cl extension
- files that are not strictly C/C++ (e.g. Android.bp, Sconscript ...)
And the following directories
- compute_kernel_writer/validation/
- tests/
- include/
- src/core/NEON/kernels/convolution/
- src/core/NEON/kernels/arm_gemm/
- src/core/NEON/kernels/arm_conv/
- data/
There will be a follow up for formatting of .cl files and the
files under tests/ and compute_kernel_writer/validation/.
Signed-off-by: Felix Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com>
Change-Id: Ib7eb1fcf4e7537b9feaefcfc15098a804a3fde0a
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10391
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Diffstat (limited to 'arm_compute')
395 files changed, 5254 insertions, 3577 deletions
diff --git a/arm_compute/Acl.h b/arm_compute/Acl.h index 1cd45d5756..3e99a00cbe 100644 --- a/arm_compute/Acl.h +++ b/arm_compute/Acl.h @@ -25,7 +25,8 @@ #define ARM_COMPUTE_ACL_H_ #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /* __cplusplus */ /* Core headers */ diff --git a/arm_compute/Acl.hpp b/arm_compute/Acl.hpp index 55e04e876d..6a9d585c14 100644 --- a/arm_compute/Acl.hpp +++ b/arm_compute/Acl.hpp @@ -75,7 +75,7 @@ struct ObjectDeleter #define OBJECT_DELETER(obj, func) \ template <> \ struct ObjectDeleter<obj> \ - \ + \ { \ static inline AclStatus Destroy(obj v) \ { \ @@ -171,7 +171,7 @@ protected: ObjectBase() = default; protected: - std::shared_ptr<T> _object{ nullptr }; /**< Library object */ + std::shared_ptr<T> _object{nullptr}; /**< Library object */ }; /** Equality operator for library object @@ -221,8 +221,7 @@ public: * @param[in] status Status returned * @param[in] msg Error message to be bound with the exception */ - Status(StatusCode status, const std::string &msg) - : _status(status), _msg(msg) + Status(StatusCode status, const std::string &msg) : _status(status), _msg(msg) { } /** Returns an explanatory exception message @@ -266,7 +265,7 @@ private: */ static inline void report_status(StatusCode status, const std::string &msg) { - if(status != StatusCode::Success) + if (status != StatusCode::Success) { throw Status(status, msg); } @@ -299,7 +298,8 @@ enum class Target /**< Available execution modes */ enum class ExecutionMode { - FastRerun = AclPreferFastRerun, /**< Prefer minimum latency in consecutive runs, might introduce higher startup times */ + FastRerun = + AclPreferFastRerun, /**< Prefer minimum latency in consecutive runs, might introduce higher startup times */ FastStart = AclPreferFastStart, /**< Prefer minimizing startup time */ }; @@ -372,8 +372,7 @@ public: * @param[in] target Target to create context for * @param[out] status Status information if requested */ - explicit Context(Target target, StatusCode *status = nullptr) - : Context(target, Options(), status) + explicit Context(Target target, StatusCode *status = nullptr) : Context(target, Options(), status) { } /** Constructor @@ -385,10 +384,11 @@ public: Context(Target target, const Options &options, StatusCode *status = nullptr) { AclContext ctx; - const auto st = detail::as_enum<StatusCode>(AclCreateContext(&ctx, detail::as_cenum<AclTarget>(target), &options.copts)); + const auto st = + detail::as_enum<StatusCode>(AclCreateContext(&ctx, detail::as_cenum<AclTarget>(target), &options.copts)); reset(ctx); report_status(st, "[Compute Library] Failed to create context"); - if(status) + if (status) { *status = st; } @@ -424,15 +424,13 @@ public: * As default options, no tuning will be performed, and the number of scheduling units will * depends on internal device discovery functionality */ - Options() - : opts{ AclTuningModeNone, 0 } {}; + Options() : opts{AclTuningModeNone, 0} {}; /** Constructor * * @param[in] mode Tuning mode to be used * @param[in] compute_units Number of scheduling units to be used */ - Options(TuningMode mode, int32_t compute_units) - : opts{ detail::as_cenum<AclTuningMode>(mode), compute_units } + Options(TuningMode mode, int32_t compute_units) : opts{detail::as_cenum<AclTuningMode>(mode), compute_units} { } @@ -448,8 +446,7 @@ public: * @param[in] ctx Context to create queue for * @param[out] status Status information if requested */ - explicit Queue(Context &ctx, StatusCode *status = nullptr) - : Queue(ctx, Options(), status) + explicit Queue(Context &ctx, StatusCode *status = nullptr) : Queue(ctx, Options(), status) { } /** Constructor @@ -466,7 +463,7 @@ public: const auto st = detail::as_enum<StatusCode>(AclCreateQueue(&queue, ctx.get(), &options.opts)); reset(queue); report_status(st, "[Compute Library] Failed to create queue!"); - if(status) + if (status) { *status = st; } @@ -508,8 +505,7 @@ public: * @param[in] shape Shape of the tensor * @param[in] data_type Data type of the tensor */ - TensorDescriptor(const std::vector<int32_t> &shape, DataType data_type) - : _shape(shape), _data_type(data_type) + TensorDescriptor(const std::vector<int32_t> &shape, DataType data_type) : _shape(shape), _data_type(data_type) { _cdesc.ndims = _shape.size(); _cdesc.shape = _shape.data(); @@ -526,7 +522,7 @@ public: _cdesc = desc; _data_type = detail::as_enum<DataType>(desc.data_type); _shape.reserve(desc.ndims); - for(int32_t d = 0; d < desc.ndims; ++d) + for (int32_t d = 0; d < desc.ndims; ++d) { _shape.emplace_back(desc.shape[d]); } @@ -552,9 +548,9 @@ public: is_same &= _data_type == other._data_type; is_same &= _shape.size() == other._shape.size(); - if(is_same) + if (is_same) { - for(uint32_t d = 0; d < _shape.size(); ++d) + for (uint32_t d = 0; d < _shape.size(); ++d) { is_same &= _shape[d] == other._shape[d]; } @@ -592,8 +588,7 @@ public: * @param[in] desc Tensor descriptor to be used * @param[out] status Status information if requested */ - Tensor(Context &ctx, const TensorDescriptor &desc, StatusCode *status = nullptr) - : Tensor(ctx, desc, true, status) + Tensor(Context &ctx, const TensorDescriptor &desc, StatusCode *status = nullptr) : Tensor(ctx, desc, true, status) { } /** Constructor @@ -609,7 +604,7 @@ public: const auto st = detail::as_enum<StatusCode>(AclCreateTensor(&tensor, ctx.get(), desc.get(), allocate)); reset(tensor); report_status(st, "[Compute Library] Failed to create tensor!"); - if(status) + if (status) { *status = st; } @@ -646,7 +641,8 @@ public: */ StatusCode import(void *handle, ImportType type) { - const auto st = detail::as_enum<StatusCode>(AclTensorImport(_object.get(), handle, detail::as_cenum<AclImportMemoryType>(type))); + const auto st = detail::as_enum<StatusCode>( + AclTensorImport(_object.get(), handle, detail::as_cenum<AclImportMemoryType>(type))); report_status(st, "[Compute Library] Failed to import external memory to tensor!"); return st; } @@ -658,7 +654,7 @@ public: */ uint64_t get_size() { - uint64_t size{ 0 }; + uint64_t size{0}; const auto st = detail::as_enum<StatusCode>(AclGetTensorSize(_object.get(), &size)); report_status(st, "[Compute Library] Failed to get the size of the tensor"); return size; @@ -692,13 +688,12 @@ public: * @param[in] tensor_ Tensor to pack * @param[in] slot_id_ Slot identification of the tensor in respect with the operator */ - PackPair(Tensor *tensor_, int32_t slot_id_) - : tensor(tensor_), slot_id(slot_id_) + PackPair(Tensor *tensor_, int32_t slot_id_) : tensor(tensor_), slot_id(slot_id_) { } - Tensor *tensor{ nullptr }; /**< Tensor object */ - int32_t slot_id{ AclSlotUnknown }; /**< Slot id in respect with the operator */ + Tensor *tensor{nullptr}; /**< Tensor object */ + int32_t slot_id{AclSlotUnknown}; /**< Slot id in respect with the operator */ }; public: @@ -713,7 +708,7 @@ public: const auto st = detail::as_enum<StatusCode>(AclCreateTensorPack(&pack, ctx.get())); reset(pack); report_status(st, "[Compute Library] Failure during tensor pack creation"); - if(status) + if (status) { *status = st; } @@ -741,7 +736,7 @@ public: std::vector<int32_t> slots(size); std::vector<AclTensor> tensors(size); int i = 0; - for(auto &p : packed) + for (auto &p : packed) { slots[i] = p.slot_id; tensors[i] = AclTensor(p.tensor); @@ -780,13 +775,17 @@ using ActivationDesc = AclActivationDescriptor; class Activation : public Operator { public: - Activation(Context &ctx, const TensorDescriptor &src, const TensorDescriptor &dst, const ActivationDesc &desc, StatusCode *status = nullptr) + Activation(Context &ctx, + const TensorDescriptor &src, + const TensorDescriptor &dst, + const ActivationDesc &desc, + StatusCode *status = nullptr) { AclOperator op; const auto st = detail::as_enum<StatusCode>(AclActivation(&op, ctx.get(), src.get(), dst.get(), desc)); reset(op); report_status(st, "[Compute Library] Failure during Activation operator creation"); - if(status) + if (status) { *status = st; } diff --git a/arm_compute/AclDescriptors.h b/arm_compute/AclDescriptors.h index a564bd2141..cdaf7c0dc8 100644 --- a/arm_compute/AclDescriptors.h +++ b/arm_compute/AclDescriptors.h @@ -25,37 +25,38 @@ #define ARM_COMPUTE_ACL_DESCRIPTORS_H_ #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /** __cplusplus */ -/**< Supported activation types */ -typedef enum -{ - AclActivationTypeNone = 0, /**< No activation */ - AclIdentity = 1, /**< Identity */ - AclLogistic = 2, /**< Logistic */ - AclTanh = 3, /**< Hyperbolic tangent */ - AclRelu = 4, /**< Rectifier */ - AclBoundedRelu = 5, /**< Upper Bounded Rectifier */ - AclLuBoundedRelu = 6, /**< Lower and Upper Bounded Rectifier */ - AclLeakyRelu = 7, /**< Leaky Rectifier */ - AclSoftRelu = 8, /**< Soft Rectifier */ - AclElu = 9, /**< Exponential Linear Unit */ - AclAbs = 10, /**< Absolute */ - AclSquare = 11, /**< Square */ - AclSqrt = 12, /**< Square root */ - AclLinear = 13, /**< Linear */ - AclHardSwish = 14, /**< Hard-swish */ -} AclActivationType; + /**< Supported activation types */ + typedef enum + { + AclActivationTypeNone = 0, /**< No activation */ + AclIdentity = 1, /**< Identity */ + AclLogistic = 2, /**< Logistic */ + AclTanh = 3, /**< Hyperbolic tangent */ + AclRelu = 4, /**< Rectifier */ + AclBoundedRelu = 5, /**< Upper Bounded Rectifier */ + AclLuBoundedRelu = 6, /**< Lower and Upper Bounded Rectifier */ + AclLeakyRelu = 7, /**< Leaky Rectifier */ + AclSoftRelu = 8, /**< Soft Rectifier */ + AclElu = 9, /**< Exponential Linear Unit */ + AclAbs = 10, /**< Absolute */ + AclSquare = 11, /**< Square */ + AclSqrt = 12, /**< Square root */ + AclLinear = 13, /**< Linear */ + AclHardSwish = 14, /**< Hard-swish */ + } AclActivationType; -/**< Activation layer descriptor */ -typedef struct -{ - AclActivationType type; /**< Activation type */ - float a; /**< Factor &alpha used by some activations */ - float b; /**< Factor &beta used by some activations */ - bool inplace; /**< Hint that src and dst tensors will be the same */ -} AclActivationDescriptor; + /**< Activation layer descriptor */ + typedef struct + { + AclActivationType type; /**< Activation type */ + float a; /**< Factor &alpha used by some activations */ + float b; /**< Factor &beta used by some activations */ + bool inplace; /**< Hint that src and dst tensors will be the same */ + } AclActivationDescriptor; #ifdef __cplusplus } #endif /** __cplusplus */ diff --git a/arm_compute/AclEntrypoints.h b/arm_compute/AclEntrypoints.h index ca3a911f5d..0d4902a3d5 100644 --- a/arm_compute/AclEntrypoints.h +++ b/arm_compute/AclEntrypoints.h @@ -27,10 +27,11 @@ #include "arm_compute/AclTypes.h" #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /** __cplusplus */ -/** Create a context object + /** Create a context object * * Context is responsible for retaining internal information and work as an aggregate service mechanism * @@ -46,11 +47,9 @@ extern "C" { * - @ref AclUnsupportedTarget if the requested target is unsupported * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclCreateContext(AclContext *ctx, - AclTarget target, - const AclContextOptions *options); + AclStatus AclCreateContext(AclContext *ctx, AclTarget target, const AclContextOptions *options); -/** Destroy a given context object + /** Destroy a given context object * * @param[in] ctx A valid context object to destroy * @@ -60,9 +59,9 @@ AclStatus AclCreateContext(AclContext *ctx, * - @ref AclSuccess if functions was completed successfully * - @ref AclInvalidArgument if the provided context is invalid */ -AclStatus AclDestroyContext(AclContext ctx); + AclStatus AclDestroyContext(AclContext ctx); -/** Create an operator queue + /** Create an operator queue * * Queue is responsible for any scheduling related activities * @@ -78,9 +77,9 @@ AclStatus AclDestroyContext(AclContext ctx); * - @ref AclUnsupportedTarget if the requested target is unsupported * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclCreateQueue(AclQueue *queue, AclContext ctx, const AclQueueOptions *options); + AclStatus AclCreateQueue(AclQueue *queue, AclContext ctx, const AclQueueOptions *options); -/** Wait until all elements on the queue have been completed + /** Wait until all elements on the queue have been completed * * @param[in] queue Queue to wait on completion * @@ -91,9 +90,9 @@ AclStatus AclCreateQueue(AclQueue *queue, AclContext ctx, const AclQueueOptions * - @ref AclInvalidArgument if the provided queue is invalid * - @ref AclRuntimeError on any other runtime related error */ -AclStatus AclQueueFinish(AclQueue queue); + AclStatus AclQueueFinish(AclQueue queue); -/** Destroy a given queue object + /** Destroy a given queue object * * @param[in] queue A valid context object to destroy * @@ -103,9 +102,9 @@ AclStatus AclQueueFinish(AclQueue queue); * - @ref AclSuccess if functions was completed successfully * - @ref AclInvalidArgument if the provided context is invalid */ -AclStatus AclDestroyQueue(AclQueue queue); + AclStatus AclDestroyQueue(AclQueue queue); -/** Create a Tensor object + /** Create a Tensor object * * Tensor is a generalized matrix construct that can represent up to ND dimensionality (where N = 6 for Compute Library) * The object holds a backing memory along-side to operate on @@ -121,9 +120,9 @@ AclStatus AclDestroyQueue(AclQueue queue); * - @ref AclUnsupportedTarget if the requested target is unsupported * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclCreateTensor(AclTensor *tensor, AclContext ctx, const AclTensorDescriptor *desc, bool allocate); + AclStatus AclCreateTensor(AclTensor *tensor, AclContext ctx, const AclTensorDescriptor *desc, bool allocate); -/** Map a tensor's backing memory to the host + /** Map a tensor's backing memory to the host * * @param[in] tensor Tensor to be mapped * @param[in, out] handle A handle to the underlying backing memory @@ -134,9 +133,9 @@ AclStatus AclCreateTensor(AclTensor *tensor, AclContext ctx, const AclTensorDesc * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclMapTensor(AclTensor tensor, void **handle); + AclStatus AclMapTensor(AclTensor tensor, void **handle); -/** Unmap the tensor's backing memory + /** Unmap the tensor's backing memory * * @param[in] tensor tensor to unmap memory from * @param[in] handle Backing memory to be unmapped @@ -147,9 +146,9 @@ AclStatus AclMapTensor(AclTensor tensor, void **handle); * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclUnmapTensor(AclTensor tensor, void *handle); + AclStatus AclUnmapTensor(AclTensor tensor, void *handle); -/** Import external memory to a given tensor object + /** Import external memory to a given tensor object * * @param[in, out] tensor Tensor to import memory to * @param[in] handle Backing memory to be imported @@ -159,9 +158,9 @@ AclStatus AclUnmapTensor(AclTensor tensor, void *handle); * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclTensorImport(AclTensor tensor, void *handle, AclImportMemoryType type); + AclStatus AclTensorImport(AclTensor tensor, void *handle, AclImportMemoryType type); -/** Destroy a given tensor object + /** Destroy a given tensor object * * @param[in,out] tensor A valid tensor object to be destroyed * @@ -171,9 +170,9 @@ AclStatus AclTensorImport(AclTensor tensor, void *handle, AclImportMemoryType ty * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if the provided tensor is invalid */ -AclStatus AclDestroyTensor(AclTensor tensor); + AclStatus AclDestroyTensor(AclTensor tensor); -/** Creates a tensor pack + /** Creates a tensor pack * * Tensor packs are used to create a collection of tensors that can be passed around for operator execution * @@ -187,9 +186,9 @@ AclStatus AclDestroyTensor(AclTensor tensor); * - @ref AclOutOfMemory if there was a failure allocating memory resources * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclCreateTensorPack(AclTensorPack *pack, AclContext ctx); + AclStatus AclCreateTensorPack(AclTensorPack *pack, AclContext ctx); -/** Add a tensor to a tensor pack + /** Add a tensor to a tensor pack * * @param[in,out] pack Pack to append a tensor to * @param[in] tensor Tensor to pack @@ -202,9 +201,9 @@ AclStatus AclCreateTensorPack(AclTensorPack *pack, AclContext ctx); * - @ref AclOutOfMemory if there was a failure allocating memory resources * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclPackTensor(AclTensorPack pack, AclTensor tensor, int32_t slot_id); + AclStatus AclPackTensor(AclTensorPack pack, AclTensor tensor, int32_t slot_id); -/** A list of tensors to a tensor pack + /** A list of tensors to a tensor pack * * @param[in,out] pack Pack to append the tensors to * @param[in] tensors Tensors to append to the pack @@ -218,9 +217,9 @@ AclStatus AclPackTensor(AclTensorPack pack, AclTensor tensor, int32_t slot_id); * - @ref AclOutOfMemory if there was a failure allocating memory resources * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclPackTensors(AclTensorPack pack, AclTensor *tensors, int32_t *slot_ids, size_t num_tensors); + AclStatus AclPackTensors(AclTensorPack pack, AclTensor *tensors, int32_t *slot_ids, size_t num_tensors); -/** Destroy a given tensor pack object + /** Destroy a given tensor pack object * * @param[in,out] pack A valid tensor pack object to destroy * @@ -230,9 +229,9 @@ AclStatus AclPackTensors(AclTensorPack pack, AclTensor *tensors, int32_t *slot_i * - @ref AclSuccess if functions was completed successfully * - @ref AclInvalidArgument if the provided context is invalid */ -AclStatus AclDestroyTensorPack(AclTensorPack pack); + AclStatus AclDestroyTensorPack(AclTensorPack pack); -/** Eager execution of a given operator on a list of inputs and outputs + /** Eager execution of a given operator on a list of inputs and outputs * * @param[in] op Operator to execute * @param[in] queue Queue to schedule the operator on @@ -247,9 +246,9 @@ AclStatus AclDestroyTensorPack(AclTensorPack pack); * - @ref AclInvalidArgument if a given argument is invalid * - @ref AclRuntimeError on any other runtime related error */ -AclStatus AclRunOperator(AclOperator op, AclQueue queue, AclTensorPack tensors); + AclStatus AclRunOperator(AclOperator op, AclQueue queue, AclTensorPack tensors); -/** Destroy a given operator object + /** Destroy a given operator object * * @param[in,out] op A valid operator object to destroy * @@ -259,7 +258,7 @@ AclStatus AclRunOperator(AclOperator op, AclQueue queue, AclTensorPack tensors); * - @ref AclSuccess if functions was completed successfully * - @ref AclInvalidArgument if the provided context is invalid */ -AclStatus AclDestroyOperator(AclOperator op); + AclStatus AclDestroyOperator(AclOperator op); #ifdef __cplusplus } #endif /* __cplusplus */ diff --git a/arm_compute/AclOpenClExt.h b/arm_compute/AclOpenClExt.h index ef80fd2443..28e918d371 100644 --- a/arm_compute/AclOpenClExt.h +++ b/arm_compute/AclOpenClExt.h @@ -37,10 +37,11 @@ #pragma GCC diagnostic pop #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /* __cplusplus */ -/** Extract the underlying OpenCL context used by a given Compute Library context object + /** Extract the underlying OpenCL context used by a given Compute Library context object * * @note @ref AclContext should be of an OpenCL backend target * @@ -49,9 +50,9 @@ extern "C" { * * @return Status code */ -AclStatus AclGetClContext(AclContext ctx, cl_context *opencl_context); + AclStatus AclGetClContext(AclContext ctx, cl_context *opencl_context); -/** Extract the underlying OpenCL device id used by a given Compute Library context object + /** Extract the underlying OpenCL device id used by a given Compute Library context object * * @note @ref AclContext should be of an OpenCL backend target * @@ -60,9 +61,9 @@ AclStatus AclGetClContext(AclContext ctx, cl_context *opencl_context); * * @return Status code */ -AclStatus AclGetClDevice(AclContext ctx, cl_device_id *opencl_device); + AclStatus AclGetClDevice(AclContext ctx, cl_device_id *opencl_device); -/** Set the underlying OpenCL context to be used by a given Compute Library context object + /** Set the underlying OpenCL context to be used by a given Compute Library context object * * @note @ref AclContext should be of an OpenCL backend target * @@ -71,9 +72,9 @@ AclStatus AclGetClDevice(AclContext ctx, cl_device_id *opencl_device); * * @return Status code */ -AclStatus AclSetClContext(AclContext ctx, cl_context opencl_context); + AclStatus AclSetClContext(AclContext ctx, cl_context opencl_context); -/** Extract the underlying OpenCL queue used by a given Compute Library queue object + /** Extract the underlying OpenCL queue used by a given Compute Library queue object * * @note @ref AclQueue should be of an OpenCL backend target * @note @ref AclQueue refcount should be 0, meaning not used by other objects @@ -83,9 +84,9 @@ AclStatus AclSetClContext(AclContext ctx, cl_context opencl_context); * * @return Status code */ -AclStatus AclGetClQueue(AclQueue queue, cl_command_queue *opencl_queue); + AclStatus AclGetClQueue(AclQueue queue, cl_command_queue *opencl_queue); -/** Set the underlying OpenCL queue to be used by a given Compute Library queue object + /** Set the underlying OpenCL queue to be used by a given Compute Library queue object * * @note @ref AclQueue should be of an OpenCL backend target * @note opecl_queue needs to be created from the same context that the AclContext that the queue will use @@ -95,16 +96,16 @@ AclStatus AclGetClQueue(AclQueue queue, cl_command_queue *opencl_queue); * * @return Status code */ -AclStatus AclSetClQueue(AclQueue queue, cl_command_queue opencl_queue); + AclStatus AclSetClQueue(AclQueue queue, cl_command_queue opencl_queue); -/** Extract the underlying OpenCL memory object by a given Compute Library tensor object + /** Extract the underlying OpenCL memory object by a given Compute Library tensor object * * @param[in] tensor A valid non-zero tensor * @param[out] opencl_mem Underlyig OpenCL memory object * * @return Status code */ -AclStatus AclGetClMem(AclTensor tensor, cl_mem *opencl_mem); + AclStatus AclGetClMem(AclTensor tensor, cl_mem *opencl_mem); #ifdef __cplusplus } diff --git a/arm_compute/AclOperators.h b/arm_compute/AclOperators.h index bfdd7b1b9b..4f6f46e9c8 100644 --- a/arm_compute/AclOperators.h +++ b/arm_compute/AclOperators.h @@ -31,10 +31,11 @@ #define ARM_COMPUTE_VALIDATE_OPERATOR_SUPPORT ((AclOperator *)(size_t)-1) #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /** __cplusplus */ -/** Create an activation operator + /** Create an activation operator * * Applies an activation function to a given tensor . * Compute Library supports a wide list of activation functions @ref AclActivationType. @@ -75,11 +76,11 @@ extern "C" { * - @ref AclUnsupportedTarget if operator for the requested target is unsupported * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclActivation(AclOperator *op, - AclContext ctx, - const AclTensorDescriptor *src, - const AclTensorDescriptor *dst, - const AclActivationDescriptor info); + AclStatus AclActivation(AclOperator *op, + AclContext ctx, + const AclTensorDescriptor *src, + const AclTensorDescriptor *dst, + const AclActivationDescriptor info); #ifdef __cplusplus } #endif /** __cplusplus */ diff --git a/arm_compute/AclTypes.h b/arm_compute/AclTypes.h index 368be1292b..9a002ad22c 100644 --- a/arm_compute/AclTypes.h +++ b/arm_compute/AclTypes.h @@ -28,185 +28,186 @@ #include <stdint.h> #ifdef __cplusplus -extern "C" { -#endif /* __cplusplus */ - -/**< Opaque Context object */ -typedef struct AclContext_ *AclContext; -/**< Opaque Queue object */ -typedef struct AclQueue_ *AclQueue; -/**< Opaque Tensor object */ -typedef struct AclTensor_ *AclTensor; -/**< Opaque Tensor pack object */ -typedef struct AclTensorPack_ *AclTensorPack; -/**< Opaque Operator object */ -typedef struct AclOperator_ *AclOperator; - -// Capabilities bitfield (Note: if multiple are enabled ComputeLibrary will pick the best possible) -typedef uint64_t AclTargetCapabilities; - -/**< Error codes returned by the public entry-points */ -typedef enum AclStatus : int32_t -{ - AclSuccess = 0, /**< Call succeeded, leading to valid state for all involved objects/data */ - AclRuntimeError = 1, /**< Call failed during execution */ - AclOutOfMemory = 2, /**< Call failed due to failure to allocate resources */ - AclUnimplemented = 3, /**< Call failed as requested capability is not implemented */ - AclUnsupportedTarget = 4, /**< Call failed as an invalid backend was requested */ - AclInvalidTarget = 5, /**< Call failed as invalid argument was passed */ - AclInvalidArgument = 6, /**< Call failed as invalid argument was passed */ - AclUnsupportedConfig = 7, /**< Call failed as configuration is unsupported */ - AclInvalidObjectState = 8, /**< Call failed as an object has invalid state */ -} AclStatus; - -/**< Supported CPU targets */ -typedef enum AclTarget -{ - AclCpu = 0, /**< Cpu target that uses SIMD extensions */ - AclGpuOcl = 1, /**< OpenCL target for GPU */ -} AclTarget; - -/** Execution mode types */ -typedef enum AclExecutionMode -{ - AclPreferFastRerun = 0, /**< Prioritize performance when multiple iterations are performed */ - AclPreferFastStart = 1, /**< Prioritize performance when a single iterations is expected to be performed */ -} AclExecutionMode; - -/** Available CPU capabilities */ -typedef enum AclCpuCapabilities +extern "C" { - AclCpuCapabilitiesAuto = 0, /**< Automatic discovery of capabilities */ - - AclCpuCapabilitiesNeon = (1 << 0), /**< Enable NEON optimized paths */ - AclCpuCapabilitiesSve = (1 << 1), /**< Enable SVE optimized paths */ - AclCpuCapabilitiesSve2 = (1 << 2), /**< Enable SVE2 optimized paths */ - // Reserve 3, 4, 5, 6 - - AclCpuCapabilitiesFp16 = (1 << 7), /**< Enable float16 data-type support */ - AclCpuCapabilitiesBf16 = (1 << 8), /**< Enable bfloat16 data-type support */ - // Reserve 9, 10, 11, 12 - - AclCpuCapabilitiesDot = (1 << 13), /**< Enable paths that use the udot/sdot instructions */ - AclCpuCapabilitiesMmlaInt8 = (1 << 14), /**< Enable paths that use the mmla integer instructions */ - AclCpuCapabilitiesMmlaFp = (1 << 15), /**< Enable paths that use the mmla float instructions */ - - AclCpuCapabilitiesAll = ~0 /**< Enable all paths */ -} AclCpuCapabilities; +#endif /* __cplusplus */ -/**< Allocator interface that can be passed to a context */ -typedef struct AclAllocator -{ - /** Allocate a block of size bytes of memory. + /**< Opaque Context object */ + typedef struct AclContext_ *AclContext; + /**< Opaque Queue object */ + typedef struct AclQueue_ *AclQueue; + /**< Opaque Tensor object */ + typedef struct AclTensor_ *AclTensor; + /**< Opaque Tensor pack object */ + typedef struct AclTensorPack_ *AclTensorPack; + /**< Opaque Operator object */ + typedef struct AclOperator_ *AclOperator; + + // Capabilities bitfield (Note: if multiple are enabled ComputeLibrary will pick the best possible) + typedef uint64_t AclTargetCapabilities; + + /**< Error codes returned by the public entry-points */ + typedef enum AclStatus : int32_t + { + AclSuccess = 0, /**< Call succeeded, leading to valid state for all involved objects/data */ + AclRuntimeError = 1, /**< Call failed during execution */ + AclOutOfMemory = 2, /**< Call failed due to failure to allocate resources */ + AclUnimplemented = 3, /**< Call failed as requested capability is not implemented */ + AclUnsupportedTarget = 4, /**< Call failed as an invalid backend was requested */ + AclInvalidTarget = 5, /**< Call failed as invalid argument was passed */ + AclInvalidArgument = 6, /**< Call failed as invalid argument was passed */ + AclUnsupportedConfig = 7, /**< Call failed as configuration is unsupported */ + AclInvalidObjectState = 8, /**< Call failed as an object has invalid state */ + } AclStatus; + + /**< Supported CPU targets */ + typedef enum AclTarget + { + AclCpu = 0, /**< Cpu target that uses SIMD extensions */ + AclGpuOcl = 1, /**< OpenCL target for GPU */ + } AclTarget; + + /** Execution mode types */ + typedef enum AclExecutionMode + { + AclPreferFastRerun = 0, /**< Prioritize performance when multiple iterations are performed */ + AclPreferFastStart = 1, /**< Prioritize performance when a single iterations is expected to be performed */ + } AclExecutionMode; + + /** Available CPU capabilities */ + typedef enum AclCpuCapabilities + { + AclCpuCapabilitiesAuto = 0, /**< Automatic discovery of capabilities */ + + AclCpuCapabilitiesNeon = (1 << 0), /**< Enable NEON optimized paths */ + AclCpuCapabilitiesSve = (1 << 1), /**< Enable SVE optimized paths */ + AclCpuCapabilitiesSve2 = (1 << 2), /**< Enable SVE2 optimized paths */ + // Reserve 3, 4, 5, 6 + + AclCpuCapabilitiesFp16 = (1 << 7), /**< Enable float16 data-type support */ + AclCpuCapabilitiesBf16 = (1 << 8), /**< Enable bfloat16 data-type support */ + // Reserve 9, 10, 11, 12 + + AclCpuCapabilitiesDot = (1 << 13), /**< Enable paths that use the udot/sdot instructions */ + AclCpuCapabilitiesMmlaInt8 = (1 << 14), /**< Enable paths that use the mmla integer instructions */ + AclCpuCapabilitiesMmlaFp = (1 << 15), /**< Enable paths that use the mmla float instructions */ + + AclCpuCapabilitiesAll = ~0 /**< Enable all paths */ + } AclCpuCapabilities; + + /**< Allocator interface that can be passed to a context */ + typedef struct AclAllocator + { + /** Allocate a block of size bytes of memory. * * @param[in] user_data User provided data that can be used by the allocator * @param[in] size Size of the allocation * * @return A pointer to the allocated block if successfull else NULL */ - void *(*alloc)(void *user_data, size_t size); - /** Release a block of size bytes of memory. + void *(*alloc)(void *user_data, size_t size); + /** Release a block of size bytes of memory. * * @param[in] user_data User provided data that can be used by the allocator * @param[in] size Size of the allocation */ - void (*free)(void *user_data, void *ptr); - /** Allocate a block of size bytes of memory. + void (*free)(void *user_data, void *ptr); + /** Allocate a block of size bytes of memory. * * @param[in] user_data User provided data that can be used by the allocator * @param[in] size Size of the allocation * * @return A pointer to the allocated block if successfull else NULL */ - void *(*aligned_alloc)(void *user_data, size_t size, size_t alignment); - /** Allocate a block of size bytes of memory. + void *(*aligned_alloc)(void *user_data, size_t size, size_t alignment); + /** Allocate a block of size bytes of memory. * * @param[in] user_data User provided data that can be used by the allocator * @param[in] size Size of the allocation */ - void (*aligned_free)(void *user_data, void *ptr); - - /**< User provided information */ - void *user_data; -} AclAllocator; - -/**< Context options */ -typedef struct AclContextOptions -{ - AclExecutionMode mode; /**< Execution mode to use */ - AclTargetCapabilities capabilities; /**< Target capabilities */ - bool enable_fast_math; /**< Allow precision loss */ - const char *kernel_config_file; /**< Kernel cofiguration file */ - int32_t max_compute_units; /**< Max compute units that can be used by a queue created from the context. + void (*aligned_free)(void *user_data, void *ptr); + + /**< User provided information */ + void *user_data; + } AclAllocator; + + /**< Context options */ + typedef struct AclContextOptions + { + AclExecutionMode mode; /**< Execution mode to use */ + AclTargetCapabilities capabilities; /**< Target capabilities */ + bool enable_fast_math; /**< Allow precision loss */ + const char *kernel_config_file; /**< Kernel cofiguration file */ + int32_t max_compute_units; /**< Max compute units that can be used by a queue created from the context. If <=0 the system will use the hw concurency insted */ - AclAllocator *allocator; /**< Allocator to be used by all the memory internally */ -} AclContextOptions; - -/**< Supported tuning modes */ -typedef enum -{ - AclTuningModeNone = 0, /**< No tuning */ - AclRapid = 1, /**< Fast tuning mode, testing a small portion of the tuning space */ - AclNormal = 2, /**< Normal tuning mode, gives a good balance between tuning mode and performance */ - AclExhaustive = 3, /**< Exhaustive tuning mode, increased tuning time but with best results */ -} AclTuningMode; - -/**< Queue options */ -typedef struct -{ - AclTuningMode mode; /**< Tuning mode */ - int32_t compute_units; /**< Compute Units that the queue will deploy */ -} AclQueueOptions; - -/**< Supported data types */ -typedef enum AclDataType -{ - AclDataTypeUnknown = 0, /**< Unknown data type */ - AclUInt8 = 1, /**< 8-bit unsigned integer */ - AclInt8 = 2, /**< 8-bit signed integer */ - AclUInt16 = 3, /**< 16-bit unsigned integer */ - AclInt16 = 4, /**< 16-bit signed integer */ - AclUint32 = 5, /**< 32-bit unsigned integer */ - AclInt32 = 6, /**< 32-bit signed integer */ - AclFloat16 = 7, /**< 16-bit floating point */ - AclBFloat16 = 8, /**< 16-bit brain floating point */ - AclFloat32 = 9, /**< 32-bit floating point */ -} AclDataType; - -/**< Supported data layouts for operations */ -typedef enum AclDataLayout -{ - AclDataLayoutUnknown = 0, /**< Unknown data layout */ - AclNhwc = 1, /**< Native, performant, Compute Library data layout */ - AclNchw = 2, /**< Data layout where width is the fastest changing dimension */ -} AclDataLayout; - -/** Type of memory to be imported */ -typedef enum AclImportMemoryType -{ - AclHostPtr = 0 /**< Host allocated memory */ -} AclImportMemoryType; - -/**< Tensor Descriptor */ -typedef struct AclTensorDescriptor -{ - int32_t ndims; /**< Number or dimensions */ - int32_t *shape; /**< Tensor Shape */ - AclDataType data_type; /**< Tensor Data type */ - int64_t *strides; /**< Strides on each dimension. Linear memory is assumed if nullptr */ - int64_t boffset; /**< Offset in terms of bytes for the first element */ -} AclTensorDescriptor; - -/**< Slot type of a tensor */ -typedef enum -{ - AclSlotUnknown = -1, - AclSrc = 0, - AclSrc0 = 0, - AclSrc1 = 1, - AclDst = 30, - AclSrcVec = 256, -} AclTensorSlot; + AclAllocator *allocator; /**< Allocator to be used by all the memory internally */ + } AclContextOptions; + + /**< Supported tuning modes */ + typedef enum + { + AclTuningModeNone = 0, /**< No tuning */ + AclRapid = 1, /**< Fast tuning mode, testing a small portion of the tuning space */ + AclNormal = 2, /**< Normal tuning mode, gives a good balance between tuning mode and performance */ + AclExhaustive = 3, /**< Exhaustive tuning mode, increased tuning time but with best results */ + } AclTuningMode; + + /**< Queue options */ + typedef struct + { + AclTuningMode mode; /**< Tuning mode */ + int32_t compute_units; /**< Compute Units that the queue will deploy */ + } AclQueueOptions; + + /**< Supported data types */ + typedef enum AclDataType + { + AclDataTypeUnknown = 0, /**< Unknown data type */ + AclUInt8 = 1, /**< 8-bit unsigned integer */ + AclInt8 = 2, /**< 8-bit signed integer */ + AclUInt16 = 3, /**< 16-bit unsigned integer */ + AclInt16 = 4, /**< 16-bit signed integer */ + AclUint32 = 5, /**< 32-bit unsigned integer */ + AclInt32 = 6, /**< 32-bit signed integer */ + AclFloat16 = 7, /**< 16-bit floating point */ + AclBFloat16 = 8, /**< 16-bit brain floating point */ + AclFloat32 = 9, /**< 32-bit floating point */ + } AclDataType; + + /**< Supported data layouts for operations */ + typedef enum AclDataLayout + { + AclDataLayoutUnknown = 0, /**< Unknown data layout */ + AclNhwc = 1, /**< Native, performant, Compute Library data layout */ + AclNchw = 2, /**< Data layout where width is the fastest changing dimension */ + } AclDataLayout; + + /** Type of memory to be imported */ + typedef enum AclImportMemoryType + { + AclHostPtr = 0 /**< Host allocated memory */ + } AclImportMemoryType; + + /**< Tensor Descriptor */ + typedef struct AclTensorDescriptor + { + int32_t ndims; /**< Number or dimensions */ + int32_t *shape; /**< Tensor Shape */ + AclDataType data_type; /**< Tensor Data type */ + int64_t *strides; /**< Strides on each dimension. Linear memory is assumed if nullptr */ + int64_t boffset; /**< Offset in terms of bytes for the first element */ + } AclTensorDescriptor; + + /**< Slot type of a tensor */ + typedef enum + { + AclSlotUnknown = -1, + AclSrc = 0, + AclSrc0 = 0, + AclSrc1 = 1, + AclDst = 30, + AclSrcVec = 256, + } AclTensorSlot; #ifdef __cplusplus } diff --git a/arm_compute/AclUtils.h b/arm_compute/AclUtils.h index ef5fa42708..61a93e6060 100644 --- a/arm_compute/AclUtils.h +++ b/arm_compute/AclUtils.h @@ -27,10 +27,11 @@ #include "arm_compute/AclTypes.h" #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /** __cplusplus */ -/** Get the size of the existing tensor in byte + /** Get the size of the existing tensor in byte * * @note The size isn't based on allocated memory, but based on information in its descriptor (dimensions, data type, etc.). * @@ -42,9 +43,9 @@ extern "C" { * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclGetTensorSize(AclTensor tensor, uint64_t *size); + AclStatus AclGetTensorSize(AclTensor tensor, uint64_t *size); -/** Get the descriptor of this tensor + /** Get the descriptor of this tensor * * @param[in] tensor A tensor in interest * @param[out] desc The descriptor of the tensor @@ -54,7 +55,7 @@ AclStatus AclGetTensorSize(AclTensor tensor, uint64_t *size); * - @ref AclSuccess if function was completed successfully * - @ref AclInvalidArgument if a given argument is invalid */ -AclStatus AclGetTensorDescriptor(AclTensor tensor, AclTensorDescriptor *desc); + AclStatus AclGetTensorDescriptor(AclTensor tensor, AclTensorDescriptor *desc); #ifdef __cplusplus } diff --git a/arm_compute/AclVersion.h b/arm_compute/AclVersion.h index 0b05a5e7dc..6eed13b924 100644 --- a/arm_compute/AclVersion.h +++ b/arm_compute/AclVersion.h @@ -25,17 +25,18 @@ #define ARM_COMPUTE_ACL_VERSION_H_ #ifdef __cplusplus -extern "C" { +extern "C" +{ #endif /* __cplusplus */ -/** Semantic versioning information */ -typedef struct AclVersion -{ - int major; /**< Major version, is increased on API incompatible changes */ - int minor; /**< Minor version, is increased on adding back-ward compatible functionality */ - int patch; /**< Patch version, is increased when doing backward compatible fixes */ - const char *build_info; /**< Build related information */ -} AclVersion; + /** Semantic versioning information */ + typedef struct AclVersion + { + int major; /**< Major version, is increased on API incompatible changes */ + int minor; /**< Minor version, is increased on adding back-ward compatible functionality */ + int patch; /**< Patch version, is increased when doing backward compatible fixes */ + const char *build_info; /**< Build related information */ + } AclVersion; /**< Major version, is increased on API incompatible changes */ #define ARM_COMPUTE_LIBRARY_VERSION_MAJOR 0 @@ -44,11 +45,11 @@ typedef struct AclVersion /**< Patch version, is increased when doing backward compatible fixes */ #define ARM_COMPUTE_LIBRARY_VERSION_PATCH 0 -/** Get library's version meta-data + /** Get library's version meta-data * * @return Version information */ -const AclVersion *AclVersionInfo(); + const AclVersion *AclVersionInfo(); #ifdef __cplusplus } diff --git a/arm_compute/core/CL/CLCompileContext.h b/arm_compute/core/CL/CLCompileContext.h index 60e0f95f83..dcd3b45670 100644 --- a/arm_compute/core/CL/CLCompileContext.h +++ b/arm_compute/core/CL/CLCompileContext.h @@ -250,8 +250,12 @@ public: * * @return The created kernel. */ - Kernel create_kernel(const std::string &kernel_name, const std::string &program_name, const std::string &program_source, - const std::string &kernel_path, const StringSet &build_options_set, bool is_binary) const; + Kernel create_kernel(const std::string &kernel_name, + const std::string &program_name, + const std::string &program_source, + const std::string &kernel_path, + const StringSet &build_options_set, + bool is_binary) const; /** Clear the library's cache of binary programs */ @@ -323,7 +327,8 @@ private: * @param[in] program_source Source of the program. * @param[in] is_binary Flag to indicate if the program source is binary. */ - const Program &load_program(const std::string &program_name, const std::string &program_source, bool is_binary) const; + const Program & + load_program(const std::string &program_name, const std::string &program_source, bool is_binary) const; /** Generates the build options given a string of user defined ones * @@ -343,11 +348,11 @@ private: */ std::string stringify_set(const StringSet &s, const std::string &kernel_path) const; - cl::Context _context; /**< Underlying CL context. */ - CLDevice _device; /**< Underlying CL device. */ + cl::Context _context; /**< Underlying CL context. */ + CLDevice _device; /**< Underlying CL device. */ mutable std::map<std::string, const Program> _programs_map; /**< Map with all already loaded program data. */ mutable std::map<std::string, cl::Program> _built_programs_map; /**< Map with all already built program data. */ - bool _is_wbsm_supported; /**< Support of worksize batch size modifier support boolean*/ + bool _is_wbsm_supported; /**< Support of worksize batch size modifier support boolean*/ }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLCOMPILECONTEXT_H */ diff --git a/arm_compute/core/CL/CLDevice.h b/arm_compute/core/CL/CLDevice.h index 5e0f86e6d9..ded6bb8493 100644 --- a/arm_compute/core/CL/CLDevice.h +++ b/arm_compute/core/CL/CLDevice.h @@ -44,8 +44,7 @@ class CLDevice : public IDevice { public: /** Default Constructor */ - CLDevice() - : _device(cl::Device()), _options() + CLDevice() : _device(cl::Device()), _options() { } @@ -53,8 +52,7 @@ public: * * @param[in] cl_device OpenCL device */ - CLDevice(const cl::Device &cl_device) - : _device(), _options() + CLDevice(const cl::Device &cl_device) : _device(), _options() { _device = cl_device; @@ -66,13 +64,13 @@ public: std::string extensions = _device.getInfo<CL_DEVICE_EXTENSIONS>(); std::istringstream iss(extensions); - for(std::string s; iss >> s;) + for (std::string s; iss >> s;) { _options.extensions.insert(s); } // SW workaround for G76 - if(_options.gpu_target == GPUTarget::G76) + if (_options.gpu_target == GPUTarget::G76) { _options.extensions.insert("cl_arm_integer_dot_product_int8"); } @@ -153,15 +151,15 @@ public: */ std::tuple<bool, std::string> is_non_uniform_workgroup_supported() const { - if(version() == CLVersion::CL30 && get_cl_non_uniform_work_group_supported(_device)) + if (version() == CLVersion::CL30 && get_cl_non_uniform_work_group_supported(_device)) { return {true, " -cl-std=CL3.0 "}; } - else if(version() == CLVersion::CL20) + else if (version() == CLVersion::CL20) { return {true, " -cl-std=CL2.0 "}; } - else if(supported("cl_arm_non_uniform_work_group_size")) + else if (supported("cl_arm_non_uniform_work_group_size")) { return {true, " -cl-arm-non-uniform-work-group-size "}; } diff --git a/arm_compute/core/CL/CLHelpers.h b/arm_compute/core/CL/CLHelpers.h index 20d93df5a1..1a639e47f9 100644 --- a/arm_compute/core/CL/CLHelpers.h +++ b/arm_compute/core/CL/CLHelpers.h @@ -179,7 +179,9 @@ bool dot8_acc_supported(const cl::Device &device); * * @return True if the configuration is supported */ -bool cl_winograd_convolution_layer_supported(const Size2D &output_tile, const Size2D &kernel_size, DataLayout data_layout); +bool cl_winograd_convolution_layer_supported(const Size2D &output_tile, + const Size2D &kernel_size, + DataLayout data_layout); /** Helper function to get the preferred native vector width size for built-in scalar types that can be put into vectors * @@ -215,7 +217,9 @@ bool image2d_from_buffer_supported(const cl::Device &device); * * @return An opencl kernel */ -cl::Kernel create_kernel(const CLCompileContext &ctx, const std::string &kernel_name, const std::set<std::string> &build_opts = std::set<std::string>()); +cl::Kernel create_kernel(const CLCompileContext &ctx, + const std::string &kernel_name, + const std::set<std::string> &build_opts = std::set<std::string>()); /** Creates a suitable LWS hint object for parallel implementations. Sets the number of WG based on the input size. * If input width is smaller than 128 we can use fewer threads than 8. diff --git a/arm_compute/core/CL/CLTypes.h b/arm_compute/core/CL/CLTypes.h index 00b7cda2e1..0f088e2b10 100644 --- a/arm_compute/core/CL/CLTypes.h +++ b/arm_compute/core/CL/CLTypes.h @@ -63,15 +63,13 @@ struct CLDeviceOptions struct CLQuantization { /** Default Constructor */ - CLQuantization() - : scale(nullptr), offset(nullptr) {}; + CLQuantization() : scale(nullptr), offset(nullptr){}; /** Constructor * * @param[in] scale OpenCL scale array * @param[in] offset OpenCL offset array */ - CLQuantization(const ICLFloatArray *scale, const ICLInt32Array *offset) - : scale(scale), offset(offset) {}; + CLQuantization(const ICLFloatArray *scale, const ICLInt32Array *offset) : scale(scale), offset(offset){}; const ICLFloatArray *scale; /**< Quantization scale array */ const ICLInt32Array *offset; /**< Quantization offset array */ diff --git a/arm_compute/core/CL/ICLArray.h b/arm_compute/core/CL/ICLArray.h index 57f842b6f9..a2b2baa5b3 100644 --- a/arm_compute/core/CL/ICLArray.h +++ b/arm_compute/core/CL/ICLArray.h @@ -40,8 +40,7 @@ public: * @param[in] max_num_values Maximum size of the array. * */ - explicit ICLArray(size_t max_num_values) - : IArray<T>(max_num_values), _mapping(nullptr) + explicit ICLArray(size_t max_num_values) : IArray<T>(max_num_values), _mapping(nullptr) { } @@ -125,5 +124,5 @@ using ICLInt16Array = ICLArray<cl_short>; using ICLInt32Array = ICLArray<cl_int>; /** Interface for OpenCL Array of floats. */ using ICLFloatArray = ICLArray<cl_float>; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_ICLARRAY_H*/ diff --git a/arm_compute/core/CL/ICLTensor.h b/arm_compute/core/CL/ICLTensor.h index 78d3757e59..8de5423762 100644 --- a/arm_compute/core/CL/ICLTensor.h +++ b/arm_compute/core/CL/ICLTensor.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_ICLTENSOR_H #define ARM_COMPUTE_ICLTENSOR_H -#include "arm_compute/core/ITensor.h" - #include "arm_compute/core/CL/CLTypes.h" +#include "arm_compute/core/ITensor.h" #include <cstdint> @@ -34,7 +33,7 @@ namespace cl { class Buffer; class CommandQueue; -} +} // namespace cl namespace arm_compute { @@ -113,5 +112,5 @@ private: }; using ICLImage = ICLTensor; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_ICLTENSOR_H */ diff --git a/arm_compute/core/CL/OpenCL.h b/arm_compute/core/CL/OpenCL.h index f86d55a9ea..a5c4e39df2 100644 --- a/arm_compute/core/CL/OpenCL.h +++ b/arm_compute/core/CL/OpenCL.h @@ -31,8 +31,8 @@ #ifndef ARM_COMPUTE_NO_EXCEPTIONS #define CL_HPP_ENABLE_EXCEPTIONS #endif // ARM_COMPUTE_NO_EXCEPTIONS -#define CL_TARGET_OPENCL_VERSION 300 -#define CL_HPP_TARGET_OPENCL_VERSION 110 +#define CL_TARGET_OPENCL_VERSION 300 +#define CL_HPP_TARGET_OPENCL_VERSION 110 #define CL_HPP_MINIMUM_OPENCL_VERSION 110 #pragma GCC diagnostic push #pragma GCC diagnostic ignored "-Weffc++" @@ -40,7 +40,7 @@ #pragma GCC diagnostic ignored "-Wunused-parameter" #if defined(__GNUG__) && __GNUG__ >= 8 #pragma GCC diagnostic ignored "-Wcatch-value" -#endif // defined(__GNUG__) && __GNUG__ >= 8 +#endif // defined(__GNUG__) && __GNUG__ >= 8 #include <CL/opencl.hpp> // include new hpp header instead of cl2.hpp #pragma GCC diagnostic pop @@ -88,8 +88,7 @@ public: */ bool load_default(); -#define DECLARE_FUNCTION_PTR(func_name) \ - std::function<decltype(func_name)> func_name##_ptr = nullptr +#define DECLARE_FUNCTION_PTR(func_name) std::function<decltype(func_name)> func_name##_ptr = nullptr DECLARE_FUNCTION_PTR(clCreateContext); DECLARE_FUNCTION_PTR(clCreateContextFromType); diff --git a/arm_compute/core/CPP/CPPTypes.h b/arm_compute/core/CPP/CPPTypes.h index e4cbd9ff9b..b080a86938 100644 --- a/arm_compute/core/CPP/CPPTypes.h +++ b/arm_compute/core/CPP/CPPTypes.h @@ -78,10 +78,10 @@ public: /* Delete move and copy constructors and assignment operator s */ - CPUInfo(CPUInfo const &) = delete; // Copy construct - CPUInfo(CPUInfo &&) = delete; // Move construct + CPUInfo(CPUInfo const &) = delete; // Copy construct + CPUInfo(CPUInfo &&) = delete; // Move construct CPUInfo &operator=(CPUInfo const &) = delete; // Copy assign - CPUInfo &operator=(CPUInfo &&) = delete; // Move assign + CPUInfo &operator=(CPUInfo &&) = delete; // Move assign /** Checks if the cpu model supports fp16. * @@ -179,9 +179,9 @@ private: /** Information about executing thread and CPU. */ struct ThreadInfo { - int thread_id{ 0 }; - int num_threads{ 1 }; - const CPUInfo *cpu_info{ nullptr }; + int thread_id{0}; + int num_threads{1}; + const CPUInfo *cpu_info{nullptr}; }; } // namespace arm_compute #endif /* ARM_COMPUTE_CPP_TYPES_H */ diff --git a/arm_compute/core/CPP/ICPPKernel.h b/arm_compute/core/CPP/ICPPKernel.h index 00a10555e3..03967a536d 100644 --- a/arm_compute/core/CPP/ICPPKernel.h +++ b/arm_compute/core/CPP/ICPPKernel.h @@ -25,9 +25,9 @@ #define ARM_COMPUTE_ICPPKERNEL_H #include "arm_compute/core/CPP/CPPTypes.h" +#include "arm_compute/core/experimental/Types.h" #include "arm_compute/core/IKernel.h" #include "arm_compute/core/Types.h" -#include "arm_compute/core/experimental/Types.h" namespace arm_compute { @@ -38,7 +38,7 @@ class ITensor; class ICPPKernel : public IKernel { public: - static constexpr size_t default_mws = 1; /* Default minimum workload size value - no impact */ + static constexpr size_t default_mws = 1; /* Default minimum workload size value - no impact */ /** Default destructor */ virtual ~ICPPKernel() = default; diff --git a/arm_compute/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.h b/arm_compute/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.h index 068b37d80c..dd91595ea6 100644 --- a/arm_compute/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.h +++ b/arm_compute/core/CPP/kernels/CPPBoxWithNonMaximaSuppressionLimitKernel.h @@ -63,8 +63,16 @@ public: * @param[out] keeps_size (Optional) Number of filtered indices per class tensor of size [num_classes]. Data types supported: U32 * @param[in] info (Optional) BoxNMSLimitInfo information. */ - void configure(const ITensor *scores_in, const ITensor *boxes_in, const ITensor *batch_splits_in, ITensor *scores_out, ITensor *boxes_out, ITensor *classes, - ITensor *batch_splits_out = nullptr, ITensor *keeps = nullptr, ITensor *keeps_size = nullptr, const BoxNMSLimitInfo info = BoxNMSLimitInfo()); + void configure(const ITensor *scores_in, + const ITensor *boxes_in, + const ITensor *batch_splits_in, + ITensor *scores_out, + ITensor *boxes_out, + ITensor *classes, + ITensor *batch_splits_out = nullptr, + ITensor *keeps = nullptr, + ITensor *keeps_size = nullptr, + const BoxNMSLimitInfo info = BoxNMSLimitInfo()); // Inherited methods overridden: void run(const Window &window, const ThreadInfo &info) override; @@ -74,9 +82,9 @@ public: void run_nmslimit(); private: - const ITensor *_scores_in; - const ITensor *_boxes_in; - const ITensor *_batch_splits_in; + const ITensor *_scores_in; + const ITensor *_boxes_in; + const ITensor *_batch_splits_in; ITensor *_scores_out; ITensor *_boxes_out; ITensor *_classes; diff --git a/arm_compute/core/CPP/kernels/CPPNonMaximumSuppressionKernel.h b/arm_compute/core/CPP/kernels/CPPNonMaximumSuppressionKernel.h index e32b5d8f7b..d1f7f8670f 100644 --- a/arm_compute/core/CPP/kernels/CPPNonMaximumSuppressionKernel.h +++ b/arm_compute/core/CPP/kernels/CPPNonMaximumSuppressionKernel.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPP_NONMAXIMUMSUPPRESSIONKERNEL_LAYER_H #define ARM_COMPUTE_CPP_NONMAXIMUMSUPPRESSIONKERNEL_LAYER_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -65,7 +64,12 @@ public: * @param[in] iou_threshold The threshold used in non maximum suppression. * */ - void configure(const ITensor *input_bboxes, const ITensor *input_scores, ITensor *output_indices, unsigned int max_output_size, const float score_threshold, const float iou_threshold); + void configure(const ITensor *input_bboxes, + const ITensor *input_scores, + ITensor *output_indices, + unsigned int max_output_size, + const float score_threshold, + const float iou_threshold); /** Static function to check if given arguments will lead to a valid configuration of @ref CPPNonMaximumSuppressionKernel * @@ -77,8 +81,12 @@ public: * @param[in] iou_threshold The threshold used in non maximum suppression. * */ - static Status validate(const ITensorInfo *input_bboxes, const ITensorInfo *input_scores, const ITensorInfo *output_indices, unsigned int max_output_size, - const float score_threshold, const float iou_threshold); + static Status validate(const ITensorInfo *input_bboxes, + const ITensorInfo *input_scores, + const ITensorInfo *output_indices, + unsigned int max_output_size, + const float score_threshold, + const float iou_threshold); // Inherited methods overridden: void run(const Window &window, const ThreadInfo &info) override; diff --git a/arm_compute/core/CPP/kernels/CPPTopKVKernel.h b/arm_compute/core/CPP/kernels/CPPTopKVKernel.h index 1245dbc14c..7326a10e2f 100644 --- a/arm_compute/core/CPP/kernels/CPPTopKVKernel.h +++ b/arm_compute/core/CPP/kernels/CPPTopKVKernel.h @@ -69,7 +69,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *predictions, const ITensorInfo *targets, ITensorInfo *output, const unsigned int k); + static Status + validate(const ITensorInfo *predictions, const ITensorInfo *targets, ITensorInfo *output, const unsigned int k); // Inherited methods overridden: void run(const Window &window, const ThreadInfo &info) override; diff --git a/arm_compute/core/Coordinates.h b/arm_compute/core/Coordinates.h index f6e1f4d282..d1240bb10a 100644 --- a/arm_compute/core/Coordinates.h +++ b/arm_compute/core/Coordinates.h @@ -42,8 +42,7 @@ public: * @param[in] coords Values to initialize the dimensions. */ template <typename... Ts> - constexpr Coordinates(Ts... coords) - : Dimensions{ coords... } + constexpr Coordinates(Ts... coords) : Dimensions{coords...} { } /** Allow instances of this class to be copy constructed */ @@ -57,5 +56,5 @@ public: /** Default destructor */ ~Coordinates() = default; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_COORDINATES_H*/ diff --git a/arm_compute/core/CoreTypes.h b/arm_compute/core/CoreTypes.h index 4a48a36651..1a9db1937c 100644 --- a/arm_compute/core/CoreTypes.h +++ b/arm_compute/core/CoreTypes.h @@ -25,6 +25,7 @@ #define ACL_ARM_COMPUTE_CORE_CORETYPES #include "arm_compute/core/Strides.h" + #include "support/Half.h" /** CoreTypes.h groups together essential small types that are used across functions */ @@ -146,9 +147,11 @@ public: * @param[in] pad_y (Optional) Padding, in elements, across y. Defaults to 0. * @param[in] round (Optional) Dimensions rounding. Defaults to @ref DimensionRoundingType::FLOOR. */ - PadStrideInfo(unsigned int stride_x = 1, unsigned int stride_y = 1, - unsigned int pad_x = 0, unsigned int pad_y = 0, - DimensionRoundingType round = DimensionRoundingType::FLOOR) + PadStrideInfo(unsigned int stride_x = 1, + unsigned int stride_y = 1, + unsigned int pad_x = 0, + unsigned int pad_y = 0, + DimensionRoundingType round = DimensionRoundingType::FLOOR) : _stride(std::make_pair(stride_x, stride_y)), _pad_left(pad_x), _pad_top(pad_y), @@ -167,9 +170,12 @@ public: * @param[in] pad_bottom Padding across y on the bottom, in elements. * @param[in] round Dimensions rounding. */ - PadStrideInfo(unsigned int stride_x, unsigned int stride_y, - unsigned int pad_left, unsigned int pad_right, - unsigned int pad_top, unsigned int pad_bottom, + PadStrideInfo(unsigned int stride_x, + unsigned int stride_y, + unsigned int pad_left, + unsigned int pad_right, + unsigned int pad_top, + unsigned int pad_bottom, DimensionRoundingType round) : _stride(std::make_pair(stride_x, stride_y)), _pad_left(pad_left), @@ -243,10 +249,10 @@ public: private: std::pair<unsigned int, unsigned int> _stride; - unsigned int _pad_left; - unsigned int _pad_top; - unsigned int _pad_right; - unsigned int _pad_bottom; + unsigned int _pad_left; + unsigned int _pad_top; + unsigned int _pad_right; + unsigned int _pad_bottom; DimensionRoundingType _round_type; }; diff --git a/arm_compute/core/Dimensions.h b/arm_compute/core/Dimensions.h index 2ebfcd7f83..bb8692d70a 100644 --- a/arm_compute/core/Dimensions.h +++ b/arm_compute/core/Dimensions.h @@ -50,8 +50,7 @@ public: * @param[in] dims Values to initialize the dimensions. */ template <typename... Ts> - explicit Dimensions(Ts... dims) - : _id{ { static_cast<T>(dims)... } }, _num_dimensions{ sizeof...(dims) } + explicit Dimensions(Ts... dims) : _id{{static_cast<T>(dims)...}}, _num_dimensions{sizeof...(dims)} { } @@ -78,7 +77,7 @@ public: ARM_COMPUTE_ERROR_ON(dimension >= num_max_dimensions); _id[dimension] = value; // Don't increase the number of dimensions if the new dimension is 1 - if(increase_dim_unit || value != 1) + if (increase_dim_unit || value != 1) { _num_dimensions = std::max(_num_dimensions, dimension + 1); } @@ -108,7 +107,7 @@ public: void increment(size_t dim, T step = 1) { ARM_COMPUTE_ERROR_ON(dim >= _num_dimensions); - if((std::numeric_limits<T>::max() - _id[dim]) >= step) + if ((std::numeric_limits<T>::max() - _id[dim]) >= step) { _id[dim] += step; } @@ -162,7 +161,7 @@ public: const size_t last = std::min(_num_dimensions, first + n); - if(last > (first + 1)) + if (last > (first + 1)) { // Collapse dimensions into the first _id[first] = std::accumulate(&_id[first], &_id[last], 1, std::multiplies<T>()); @@ -196,7 +195,7 @@ public: void remove(size_t idx) { ARM_COMPUTE_ERROR_ON(_num_dimensions < 1); - if(idx >= _num_dimensions) + if (idx >= _num_dimensions) { return; } @@ -262,7 +261,7 @@ protected: ~Dimensions() = default; std::array<T, num_max_dimensions> _id; - size_t _num_dimensions{ 0 }; + size_t _num_dimensions{0}; }; /** Check that given dimensions are equal. @@ -289,5 +288,5 @@ inline bool operator!=(const Dimensions<T> &lhs, const Dimensions<T> &rhs) { return !(lhs == rhs); } -} +} // namespace arm_compute #endif /*ARM_COMPUTE_DIMENSIONS_H*/ diff --git a/arm_compute/core/Error.h b/arm_compute/core/Error.h index 0854f2c527..7a7033805a 100644 --- a/arm_compute/core/Error.h +++ b/arm_compute/core/Error.h @@ -53,8 +53,7 @@ class Status { public: /** Default Constructor **/ - Status() - : _code(ErrorCode::OK), _error_description(" ") + Status() : _code(ErrorCode::OK), _error_description(" ") { } /** Default Constructor @@ -101,7 +100,7 @@ public: /** Throws a runtime exception in case it contains a valid error status */ void throw_if_error() const { - if(!bool(*this)) + if (!bool(*this)) { internal_throw_on_error(); } @@ -141,7 +140,7 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] err Error status */ [[noreturn]] void throw_error(Status err); -} +} // namespace arm_compute /** To avoid unused variables warnings * * This is useful if for example a variable is only used @@ -156,7 +155,8 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] error_code Error code. * @param[in] msg Message to encapsulate. */ -#define ARM_COMPUTE_CREATE_ERROR(error_code, msg) arm_compute::create_error_msg(error_code, __func__, __FILE__, __LINE__, msg) +#define ARM_COMPUTE_CREATE_ERROR(error_code, msg) \ + arm_compute::create_error_msg(error_code, __func__, __FILE__, __LINE__, msg) /** Creates an error on location with a given message * @@ -166,7 +166,8 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] line Line in which the error occurred. * @param[in] msg Message to display before abandoning. */ -#define ARM_COMPUTE_CREATE_ERROR_LOC(error_code, func, file, line, msg) arm_compute::create_error_msg(error_code, func, file, line, msg) +#define ARM_COMPUTE_CREATE_ERROR_LOC(error_code, func, file, line, msg) \ + arm_compute::create_error_msg(error_code, func, file, line, msg) /** Creates an error on location with a given message. Accepts a message format * and a variable list of arguments matching the format description. @@ -178,14 +179,14 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] msg Error description message format. * @param[in] ... List of arguments matching the format description. */ -#define ARM_COMPUTE_CREATE_ERROR_LOC_VAR(error_code, func, file, line, msg, ...) \ - do \ - { \ - std::array<char, 512> out{ 0 }; \ - int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ - snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ - arm_compute::create_error(error_code, std::string(out.data())); \ - } while(false) +#define ARM_COMPUTE_CREATE_ERROR_LOC_VAR(error_code, func, file, line, msg, ...) \ + do \ + { \ + std::array<char, 512> out{0}; \ + int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ + snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ + arm_compute::create_error(error_code, std::string(out.data())); \ + } while (false) /** An error is returned with the given description. * @@ -195,7 +196,7 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file do \ { \ return ARM_COMPUTE_CREATE_ERROR(arm_compute::ErrorCode::RUNTIME_ERROR, __VA_ARGS__); \ - } while(false) + } while (false) /** Checks if a status contains an error and returns it * @@ -205,18 +206,17 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file do \ { \ const auto s = status; \ - if(!bool(s)) \ + if (!bool(s)) \ { \ return s; \ } \ - } while(false) + } while (false) /** Checks if an error value is valid if not throws an exception with the error * * @param[in] error Error value to check. */ -#define ARM_COMPUTE_THROW_ON_ERROR(error) \ - error.throw_if_error(); +#define ARM_COMPUTE_THROW_ON_ERROR(error) error.throw_if_error(); /** If the condition is true, an error is returned. Accepts a message format * and a variable list of arguments matching the format description. @@ -228,28 +228,29 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_RETURN_ERROR_ON_MSG_VAR(cond, msg, ...) \ do \ { \ - if(cond) \ + if (cond) \ { \ - std::array<char, 512> out{ 0 }; \ + std::array<char, 512> out{0}; \ int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", __func__, __FILE__, __LINE__); \ snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ return arm_compute::create_error(arm_compute::ErrorCode::RUNTIME_ERROR, std::string(out.data())); \ } \ - } while(false) + } while (false) /** If the condition is true, an error is returned * * @param[in] cond Condition to evaluate. * @param[in] msg Error description message */ -#define ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, msg) \ - do \ - { \ - if(cond) \ - { \ - return arm_compute::create_error_msg(arm_compute::ErrorCode::RUNTIME_ERROR, __func__, __FILE__, __LINE__, msg); \ - } \ - } while(false) +#define ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, msg) \ + do \ + { \ + if (cond) \ + { \ + return arm_compute::create_error_msg(arm_compute::ErrorCode::RUNTIME_ERROR, __func__, __FILE__, __LINE__, \ + msg); \ + } \ + } while (false) /** If the condition is true, an error is thrown. Accepts a message format * and a variable list of arguments matching the format description. @@ -261,17 +262,17 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] msg Error description message format. * @param[in] ... List of arguments matching the format description. */ -#define ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(cond, func, file, line, msg, ...) \ - do \ - { \ - if(cond) \ - { \ - std::array<char, 512> out{ 0 }; \ - int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ - snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ - return arm_compute::create_error(ErrorCode::RUNTIME_ERROR, std::string(out.data())); \ - } \ - } while(false) +#define ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(cond, func, file, line, msg, ...) \ + do \ + { \ + if (cond) \ + { \ + std::array<char, 512> out{0}; \ + int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ + snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ + return arm_compute::create_error(ErrorCode::RUNTIME_ERROR, std::string(out.data())); \ + } \ + } while (false) /** If the condition is true, an error is thrown. * @@ -284,18 +285,17 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(cond, func, file, line, msg) \ do \ { \ - if(cond) \ + if (cond) \ { \ return arm_compute::create_error_msg(ErrorCode::RUNTIME_ERROR, func, file, line, msg); \ } \ - } while(false) + } while (false) /** If the condition is true, an error is returned * * @param[in] cond Condition to evaluate */ -#define ARM_COMPUTE_RETURN_ERROR_ON(cond) \ - ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, #cond) +#define ARM_COMPUTE_RETURN_ERROR_ON(cond) ARM_COMPUTE_RETURN_ERROR_ON_MSG(cond, #cond) /** If the condition is true, an error is returned * @@ -314,11 +314,12 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] line Line in which the error occurred. * @param[in] msg Message to display. */ -#define ARM_COMPUTE_THROW_ERROR(func, file, line, msg) \ - do \ - { \ - arm_compute::throw_error(arm_compute::create_error_msg(arm_compute::ErrorCode::RUNTIME_ERROR, func, file, line, msg)); \ - } while(false) +#define ARM_COMPUTE_THROW_ERROR(func, file, line, msg) \ + do \ + { \ + arm_compute::throw_error( \ + arm_compute::create_error_msg(arm_compute::ErrorCode::RUNTIME_ERROR, func, file, line, msg)); \ + } while (false) /** Print the given message then throw an std::runtime_error. Accepts a message format * and a variable list of arguments matching the format description. @@ -332,11 +333,11 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_THROW_ERROR_VAR(func, file, line, msg, ...) \ do \ { \ - std::array<char, 512> out{ 0 }; \ - int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ + std::array<char, 512> out{0}; \ + int offset = snprintf(out.data(), out.size(), "in %s %s:%d: ", func, file, line); \ snprintf(out.data() + offset, out.size() - offset, msg, __VA_ARGS__); \ arm_compute::throw_error(arm_compute::Status(arm_compute::ErrorCode::RUNTIME_ERROR, std::string(out.data()))); \ - } while(false) + } while (false) /** Print the given message then throw an std::runtime_error. Accepts a message format * and a variable list of arguments matching the format description. @@ -361,7 +362,8 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] msg Error description message format. * @param[in] ... List of arguments matching the format description. */ -#define ARM_COMPUTE_ERROR_LOC_VAR(func, file, line, msg, ...) ARM_COMPUTE_THROW_ERROR_VAR(func, file, line, msg, __VA_ARGS__) // NOLINT +#define ARM_COMPUTE_ERROR_LOC_VAR(func, file, line, msg, ...) \ + ARM_COMPUTE_THROW_ERROR_VAR(func, file, line, msg, __VA_ARGS__) // NOLINT /** Print the given message then throw an std::runtime_error. * @@ -380,11 +382,11 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_EXIT_ON_MSG(cond, msg) \ do \ { \ - if(cond) \ + if (cond) \ { \ ARM_COMPUTE_ERROR(msg); \ } \ - } while(false) + } while (false) /** If the condition is true, the given message is printed and program exits. Accepts a message format * and a variable list of arguments matching the format description. @@ -396,27 +398,25 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_EXIT_ON_MSG_VAR(cond, msg, ...) \ do \ { \ - if(cond) \ + if (cond) \ { \ ARM_COMPUTE_ERROR_VAR(msg, __VA_ARGS__); \ } \ - } while(false) + } while (false) #ifdef ARM_COMPUTE_ASSERTS_ENABLED /** Checks if a status value is valid if not throws an exception with the error * * @param[in] status Status value to check. */ -#define ARM_COMPUTE_ERROR_THROW_ON(status) \ - status.throw_if_error() +#define ARM_COMPUTE_ERROR_THROW_ON(status) status.throw_if_error() /** If the condition is true, the given message is printed and an exception is thrown * * @param[in] cond Condition to evaluate. * @param[in] msg Message to display. */ -#define ARM_COMPUTE_ERROR_ON_MSG(cond, msg) \ - ARM_COMPUTE_EXIT_ON_MSG(cond, msg) +#define ARM_COMPUTE_ERROR_ON_MSG(cond, msg) ARM_COMPUTE_EXIT_ON_MSG(cond, msg) /** If the condition is true, the given message is printed and an exception is thrown. Accepts a message format * and a variable list of arguments matching the format description. @@ -425,8 +425,7 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * @param[in] msg Error description message format. * @param[in] ... List of arguments matching the format description. */ -#define ARM_COMPUTE_ERROR_ON_MSG_VAR(cond, msg, ...) \ - ARM_COMPUTE_EXIT_ON_MSG_VAR(cond, msg, __VA_ARGS__) +#define ARM_COMPUTE_ERROR_ON_MSG_VAR(cond, msg, ...) ARM_COMPUTE_EXIT_ON_MSG_VAR(cond, msg, __VA_ARGS__) /** If the condition is true, the given message is printed and an exception is thrown. * @@ -439,11 +438,11 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file #define ARM_COMPUTE_ERROR_ON_LOC_MSG(cond, func, file, line, ...) \ do \ { \ - if(cond) \ + if (cond) \ { \ ARM_COMPUTE_ERROR_LOC_VAR(func, file, line, __VA_ARGS__); \ } \ - } while(false) + } while (false) /** If the condition is true, the given message is printed and an exception is thrown, otherwise value is returned * @@ -464,8 +463,7 @@ Status create_error_msg(ErrorCode error_code, const char *func, const char *file * * @param[in] cond Condition to evaluate. */ -#define ARM_COMPUTE_ERROR_ON(cond) \ - ARM_COMPUTE_ERROR_ON_MSG(cond, #cond) +#define ARM_COMPUTE_ERROR_ON(cond) ARM_COMPUTE_ERROR_ON_MSG(cond, #cond) /** If the condition is true then an error message is printed and an exception thrown * diff --git a/arm_compute/core/Helpers.h b/arm_compute/core/Helpers.h index f19e1e12e0..960201510a 100644 --- a/arm_compute/core/Helpers.h +++ b/arm_compute/core/Helpers.h @@ -96,7 +96,6 @@ public: void reset(size_t dimension); private: - /** Initialize a container iterator for the tensor with the specified number of dimensions, stride, buffer pointer and window. * * @param[in] num_dims The number of dimensions. @@ -112,8 +111,7 @@ private: class Dimension { public: - constexpr Dimension() - : _dim_start(0), _stride(0) + constexpr Dimension() : _dim_start(0), _stride(0) { } @@ -133,7 +131,7 @@ private: * @param[in,out] iterators Tensor iterators which will be updated by this function before calling lambda_function. */ template <typename L, typename... Ts> -inline void execute_window_loop(const Window &w, L &&lambda_function, Ts &&... iterators); +inline void execute_window_loop(const Window &w, L &&lambda_function, Ts &&...iterators); /** Permutes given Dimensions according to a permutation vector * @@ -146,7 +144,7 @@ template <typename T> inline void permute(Dimensions<T> &dimensions, const PermutationVector &perm) { auto dimensions_copy = utility::make_array<Dimensions<T>::num_max_dimensions>(dimensions.begin(), dimensions.end()); - for(unsigned int i = 0; i < perm.num_dimensions(); ++i) + for (unsigned int i = 0; i < perm.num_dimensions(); ++i) { T dimension_val = (perm[i] < dimensions.num_dimensions()) ? dimensions_copy[perm[i]] : 0; dimensions.set(i, dimension_val); @@ -163,7 +161,7 @@ inline void permute(Dimensions<T> &dimensions, const PermutationVector &perm) inline void permute(TensorShape &shape, const PermutationVector &perm) { TensorShape shape_copy = shape; - for(unsigned int i = 0; i < perm.num_dimensions(); ++i) + for (unsigned int i = 0; i < perm.num_dimensions(); ++i) { size_t dimension_val = (perm[i] < shape.num_dimensions()) ? shape_copy[perm[i]] : 1; shape.set(i, dimension_val, false, false); // Avoid changes in _num_dimension @@ -180,8 +178,11 @@ inline void permute(TensorShape &shape, const PermutationVector &perm) * * @return The corresponding valid region */ -ValidRegion calculate_valid_region_scale(const ITensorInfo &src_info, const TensorShape &dst_shape, - InterpolationPolicy interpolate_policy, SamplingPolicy sampling_policy, bool border_undefined); +ValidRegion calculate_valid_region_scale(const ITensorInfo &src_info, + const TensorShape &dst_shape, + InterpolationPolicy interpolate_policy, + SamplingPolicy sampling_policy, + bool border_undefined); /** Convert a linear index into n-dimensional coordinates. * @@ -224,7 +225,8 @@ const std::map<DataLayout, std::vector<DataLayoutDimension>> &get_layout_map(); * * @return The int conversion of the requested data layout index. */ -inline size_t get_data_layout_dimension_index(const DataLayout &data_layout, const DataLayoutDimension &data_layout_dimension); +inline size_t get_data_layout_dimension_index(const DataLayout &data_layout, + const DataLayoutDimension &data_layout_dimension); /** Get the DataLayoutDimension of a given index and layout. * @@ -245,10 +247,17 @@ inline DataLayoutDimension get_index_data_layout_dimension(const DataLayout &dat * * @return the number of output tiles along the x and y directions of size "output_tile_size" */ -inline Size2D compute_winograd_convolution_tiles(const Size2D &in_dims, const Size2D &kernel_size, const Size2D &output_tile_size, const PadStrideInfo &conv_info) +inline Size2D compute_winograd_convolution_tiles(const Size2D &in_dims, + const Size2D &kernel_size, + const Size2D &output_tile_size, + const PadStrideInfo &conv_info) { - int num_tiles_x = std::ceil((in_dims.width - (kernel_size.width - 1) + conv_info.pad_left() + conv_info.pad_right()) / static_cast<float>(output_tile_size.width)); - int num_tiles_y = std::ceil((in_dims.height - (kernel_size.height - 1) + conv_info.pad_top() + conv_info.pad_bottom()) / static_cast<float>(output_tile_size.height)); + int num_tiles_x = + std::ceil((in_dims.width - (kernel_size.width - 1) + conv_info.pad_left() + conv_info.pad_right()) / + static_cast<float>(output_tile_size.width)); + int num_tiles_y = + std::ceil((in_dims.height - (kernel_size.height - 1) + conv_info.pad_top() + conv_info.pad_bottom()) / + static_cast<float>(output_tile_size.height)); // Clamp in case we provide paddings but we have 1D convolution num_tiles_x = std::min(num_tiles_x, static_cast<int>(in_dims.width)); @@ -277,7 +286,7 @@ inline T wrap_around(T x, T m) */ inline Coordinates &convert_negative_axis(Coordinates &coords, int max_value) { - for(unsigned int i = 0; i < coords.num_dimensions(); ++i) + for (unsigned int i = 0; i < coords.num_dimensions(); ++i) { coords[i] = wrap_around(coords[i], max_value); } diff --git a/arm_compute/core/Helpers.inl b/arm_compute/core/Helpers.inl index ff902bba20..60a21e9418 100644 --- a/arm_compute/core/Helpers.inl +++ b/arm_compute/core/Helpers.inl @@ -32,12 +32,9 @@ template <size_t dimension> struct IncrementIterators { template <typename T, typename... Ts> - static void unroll(T &&it, Ts &&... iterators) + static void unroll(T &&it, Ts &&...iterators) { - auto increment = [](T && it) - { - it.increment(dimension); - }; + auto increment = [](T &&it) { it.increment(dimension); }; utility::for_each(increment, std::forward<T>(it), std::forward<Ts>(iterators)...); } static void unroll() @@ -50,14 +47,14 @@ template <size_t dim> struct ForEachDimension { template <typename L, typename... Ts> - static void unroll(const Window &w, Coordinates &id, L &&lambda_function, Ts &&... iterators) + static void unroll(const Window &w, Coordinates &id, L &&lambda_function, Ts &&...iterators) { const auto &d = w[dim - 1]; - for(auto v = d.start(); v < d.end(); v += d.step(), IncrementIterators < dim - 1 >::unroll(iterators...)) + for (auto v = d.start(); v < d.end(); v += d.step(), IncrementIterators<dim - 1>::unroll(iterators...)) { id.set(dim - 1, v); - ForEachDimension < dim - 1 >::unroll(w, id, lambda_function, iterators...); + ForEachDimension<dim - 1>::unroll(w, id, lambda_function, iterators...); } } }; @@ -66,7 +63,7 @@ template <> struct ForEachDimension<0> { template <typename L, typename... Ts> - static void unroll(const Window &w, Coordinates &id, L &&lambda_function, Ts &&... iterators) + static void unroll(const Window &w, Coordinates &id, L &&lambda_function, Ts &&...iterators) { ARM_COMPUTE_UNUSED(w, iterators...); lambda_function(id); @@ -74,31 +71,31 @@ struct ForEachDimension<0> }; template <typename L, typename... Ts> -inline void execute_window_loop(const Window &w, L &&lambda_function, Ts &&... iterators) +inline void execute_window_loop(const Window &w, L &&lambda_function, Ts &&...iterators) { w.validate(); - for(unsigned int i = 0; i < Coordinates::num_max_dimensions; ++i) + for (unsigned int i = 0; i < Coordinates::num_max_dimensions; ++i) { ARM_COMPUTE_ERROR_ON(w[i].step() == 0); } Coordinates id; - ForEachDimension<Coordinates::num_max_dimensions>::unroll(w, id, std::forward<L>(lambda_function), std::forward<Ts>(iterators)...); + ForEachDimension<Coordinates::num_max_dimensions>::unroll(w, id, std::forward<L>(lambda_function), + std::forward<Ts>(iterators)...); } -inline constexpr Iterator::Iterator() - : _ptr(nullptr), _dims() +inline constexpr Iterator::Iterator() : _ptr(nullptr), _dims() { } -inline Iterator::Iterator(const ITensor *tensor, const Window &win) - : Iterator() +inline Iterator::Iterator(const ITensor *tensor, const Window &win) : Iterator() { ARM_COMPUTE_ERROR_ON(tensor == nullptr); ARM_COMPUTE_ERROR_ON(tensor->info() == nullptr); - initialize(tensor->info()->num_dimensions(), tensor->info()->strides_in_bytes(), tensor->buffer(), tensor->info()->offset_first_element_in_bytes(), win); + initialize(tensor->info()->num_dimensions(), tensor->info()->strides_in_bytes(), tensor->buffer(), + tensor->info()->offset_first_element_in_bytes(), win); } inline Iterator::Iterator(size_t num_dims, const Strides &strides, uint8_t *buffer, size_t offset, const Window &win) @@ -107,21 +104,22 @@ inline Iterator::Iterator(size_t num_dims, const Strides &strides, uint8_t *buff initialize(num_dims, strides, buffer, offset, win); } -inline void Iterator::initialize(size_t num_dims, const Strides &strides, uint8_t *buffer, size_t offset, const Window &win) +inline void +Iterator::initialize(size_t num_dims, const Strides &strides, uint8_t *buffer, size_t offset, const Window &win) { ARM_COMPUTE_ERROR_ON(buffer == nullptr); _ptr = buffer + offset; //Initialize the stride for each dimension and calculate the position of the first element of the iteration: - for(unsigned int n = 0; n < num_dims; ++n) + for (unsigned int n = 0; n < num_dims; ++n) { _dims[n]._stride = win[n].step() * strides[n]; std::get<0>(_dims)._dim_start += static_cast<size_t>(strides[n]) * win[n].start(); } //Copy the starting point to all the dimensions: - for(unsigned int n = 1; n < Coordinates::num_max_dimensions; ++n) + for (unsigned int n = 1; n < Coordinates::num_max_dimensions; ++n) { _dims[n]._dim_start = std::get<0>(_dims)._dim_start; } @@ -135,7 +133,7 @@ inline void Iterator::increment(const size_t dimension) _dims[dimension]._dim_start += _dims[dimension]._stride; - for(unsigned int n = 0; n < dimension; ++n) + for (unsigned int n = 0; n < dimension; ++n) { _dims[n]._dim_start = _dims[dimension]._dim_start; } @@ -157,7 +155,7 @@ inline void Iterator::reset(const size_t dimension) _dims[dimension]._dim_start = _dims[dimension + 1]._dim_start; - for(unsigned int n = 0; n < dimension; ++n) + for (unsigned int n = 0; n < dimension; ++n) { _dims[n]._dim_start = _dims[dimension]._dim_start; } @@ -170,9 +168,9 @@ inline Coordinates index2coords(const TensorShape &shape, int index) ARM_COMPUTE_ERROR_ON_MSG(index < 0 || index >= num_elements, "Index has to be in [0, num_elements]!"); ARM_COMPUTE_ERROR_ON_MSG(num_elements == 0, "Cannot create coordinate from empty shape!"); - Coordinates coord{ 0 }; + Coordinates coord{0}; - for(int d = shape.num_dimensions() - 1; d >= 0; --d) + for (int d = shape.num_dimensions() - 1; d >= 0; --d) { num_elements /= shape[d]; coord.set(d, index / num_elements); @@ -191,7 +189,7 @@ inline int coords2index(const TensorShape &shape, const Coordinates &coord) int index = 0; int stride = 1; - for(unsigned int d = 0; d < coord.num_dimensions(); ++d) + for (unsigned int d = 0; d < coord.num_dimensions(); ++d) { index += coord[d] * stride; stride *= shape[d]; @@ -200,9 +198,11 @@ inline int coords2index(const TensorShape &shape, const Coordinates &coord) return index; } -inline size_t get_data_layout_dimension_index(const DataLayout &data_layout, const DataLayoutDimension &data_layout_dimension) +inline size_t get_data_layout_dimension_index(const DataLayout &data_layout, + const DataLayoutDimension &data_layout_dimension) { - ARM_COMPUTE_ERROR_ON_MSG(data_layout == DataLayout::UNKNOWN, "Cannot retrieve the dimension index for an unknown layout!"); + ARM_COMPUTE_ERROR_ON_MSG(data_layout == DataLayout::UNKNOWN, + "Cannot retrieve the dimension index for an unknown layout!"); const auto &dims = get_layout_map().at(data_layout); const auto &it = std::find(dims.cbegin(), dims.cend(), data_layout_dimension); ARM_COMPUTE_ERROR_ON_MSG(it == dims.cend(), "Invalid dimension for the given layout."); @@ -211,7 +211,8 @@ inline size_t get_data_layout_dimension_index(const DataLayout &data_layout, con inline DataLayoutDimension get_index_data_layout_dimension(const DataLayout &data_layout, const size_t index) { - ARM_COMPUTE_ERROR_ON_MSG(data_layout == DataLayout::UNKNOWN, "Cannot retrieve the layout dimension for an unknown layout!"); + ARM_COMPUTE_ERROR_ON_MSG(data_layout == DataLayout::UNKNOWN, + "Cannot retrieve the layout dimension for an unknown layout!"); const auto &dims = get_layout_map().at(data_layout); ARM_COMPUTE_ERROR_ON_MSG(index >= dims.size(), "Invalid index for the given layout."); return dims[index]; diff --git a/arm_compute/core/IAccessWindow.h b/arm_compute/core/IAccessWindow.h index 880f6d6b27..9c9fb90915 100644 --- a/arm_compute/core/IAccessWindow.h +++ b/arm_compute/core/IAccessWindow.h @@ -100,7 +100,10 @@ public: * @return a valid region. * */ - virtual ValidRegion compute_valid_region(const Window &window, ValidRegion input_valid_region, bool border_undefined, BorderSize border_size) const = 0; + virtual ValidRegion compute_valid_region(const Window &window, + ValidRegion input_valid_region, + bool border_undefined, + BorderSize border_size) const = 0; }; /** Implementation of a rectangular access pattern. */ @@ -161,7 +164,10 @@ public: * @param[in] border_undefined (Optional) Undefined borders are excluded from the valid region. * @param[in] border_size (Optional) Size of the border around the XY-plane of the tensor. */ - void set_valid_region(const Window &window, const ValidRegion &input_valid_region, bool border_undefined = false, const BorderSize &border_size = BorderSize(0)); + void set_valid_region(const Window &window, + const ValidRegion &input_valid_region, + bool border_undefined = false, + const BorderSize &border_size = BorderSize(0)); /** Compute the valid region based on access pattern, valid region of the inputs and border mode. * @@ -189,7 +195,10 @@ public: * @return a valid region. * */ - ValidRegion compute_valid_region(const Window &window, ValidRegion input_valid_region, bool border_undefined, BorderSize border_size) const override; + ValidRegion compute_valid_region(const Window &window, + ValidRegion input_valid_region, + bool border_undefined, + BorderSize border_size) const override; bool update_window_if_needed(Window &window) const override; bool update_padding_if_needed(const Window &window) override; diff --git a/arm_compute/core/IArray.h b/arm_compute/core/IArray.h index 6edbc1d5d5..3471fc9a86 100644 --- a/arm_compute/core/IArray.h +++ b/arm_compute/core/IArray.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_IARRAY_H #include "arm_compute/core/Error.h" + #include <cstddef> #include <cstdint> @@ -36,14 +37,12 @@ class IArray { public: /** Default constructor */ - IArray() - : _num_values(0), _max_size(0) {}; + IArray() : _num_values(0), _max_size(0){}; /** Constructor: initializes an array which can contain up to max_num_points values * * @param[in] max_num_values Maximum number of values the array will be able to stored */ - IArray(size_t max_num_values) - : _num_values(0), _max_size(max_num_values) + IArray(size_t max_num_values) : _num_values(0), _max_size(max_num_values) { } /** Maximum number of values which can be stored in this array @@ -73,7 +72,7 @@ public: bool push_back(const T &val) { ARM_COMPUTE_ERROR_ON(0 == _max_size); - if(_num_values >= max_num_values()) + if (_num_values >= max_num_values()) { _num_values = max_num_values() + 1; return false; @@ -142,5 +141,5 @@ using IInt16Array = IArray<int16_t>; using IInt32Array = IArray<int32_t>; /** Interface for Array of floats. */ using IFloatArray = IArray<float>; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_IARRAY_H */ diff --git a/arm_compute/core/IKernel.h b/arm_compute/core/IKernel.h index 98fd18cc91..403a2c724e 100644 --- a/arm_compute/core/IKernel.h +++ b/arm_compute/core/IKernel.h @@ -73,5 +73,5 @@ protected: private: Window _window; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_IKERNEL_H */ diff --git a/arm_compute/core/ITensor.h b/arm_compute/core/ITensor.h index 32b93576bd..aad8313261 100644 --- a/arm_compute/core/ITensor.h +++ b/arm_compute/core/ITensor.h @@ -94,9 +94,9 @@ public: void mark_as_used() const; private: - mutable bool _is_used = { true }; /**< Flag that marks if the tensor is used or not */ + mutable bool _is_used = {true}; /**< Flag that marks if the tensor is used or not */ }; using IImage = ITensor; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_ITENSOR_H */ diff --git a/arm_compute/core/ITensorInfo.h b/arm_compute/core/ITensorInfo.h index e7c0b182c6..c42f4b57a1 100644 --- a/arm_compute/core/ITensorInfo.h +++ b/arm_compute/core/ITensorInfo.h @@ -29,6 +29,7 @@ #include "arm_compute/core/TensorShape.h" #include "arm_compute/core/Types.h" #include "arm_compute/core/utils/misc/Utility.h" + #include "support/ICloneable.h" #include <cstddef> @@ -328,23 +329,23 @@ public: * not broadcast compatible. */ template <typename... Infos> - static std::pair<TensorShape, ValidRegion> broadcast_shape_and_valid_region(const Infos &... infos) + static std::pair<TensorShape, ValidRegion> broadcast_shape_and_valid_region(const Infos &...infos) { TensorShape bc_shape = TensorShape::broadcast_shape(infos.tensor_shape()...); - ValidRegion bc_valid_region{ Coordinates(), bc_shape }; + ValidRegion bc_valid_region{Coordinates(), bc_shape}; - auto broadcast_valid_region = [&bc_valid_region](const ITensorInfo & info) + auto broadcast_valid_region = [&bc_valid_region](const ITensorInfo &info) { - if(info.num_dimensions() != 0) + if (info.num_dimensions() != 0) { - for(size_t d = 0; d < bc_valid_region.shape.num_dimensions(); ++d) + for (size_t d = 0; d < bc_valid_region.shape.num_dimensions(); ++d) { const bool is_broadcast = (info.tensor_shape()[d] == 1); const int anchor_max = std::max(bc_valid_region.anchor[d], info.valid_region().anchor[d]); const size_t valid_min = std::min(bc_valid_region.shape[d], info.valid_region().shape[d]); - if(!is_broadcast || (valid_min == 0)) + if (!is_broadcast || (valid_min == 0)) { bc_valid_region.anchor.set(d, anchor_max); bc_valid_region.shape.set(d, valid_min); diff --git a/arm_compute/core/ITensorPack.h b/arm_compute/core/ITensorPack.h index 17b7241862..f456c50769 100644 --- a/arm_compute/core/ITensorPack.h +++ b/arm_compute/core/ITensorPack.h @@ -42,18 +42,16 @@ public: struct PackElement { PackElement() = default; - PackElement(int id, ITensor *tensor) - : id(id), tensor(tensor), ctensor(nullptr) + PackElement(int id, ITensor *tensor) : id(id), tensor(tensor), ctensor(nullptr) { } - PackElement(int id, const ITensor *ctensor) - : id(id), tensor(nullptr), ctensor(ctensor) + PackElement(int id, const ITensor *ctensor) : id(id), tensor(nullptr), ctensor(ctensor) { } - int id{ -1 }; - ITensor *tensor{ nullptr }; - const ITensor *ctensor{ nullptr }; + int id{-1}; + ITensor *tensor{nullptr}; + const ITensor *ctensor{nullptr}; }; public: diff --git a/arm_compute/core/KernelDescriptors.h b/arm_compute/core/KernelDescriptors.h index 2bf5dee18c..168a06a55c 100644 --- a/arm_compute/core/KernelDescriptors.h +++ b/arm_compute/core/KernelDescriptors.h @@ -33,24 +33,24 @@ namespace arm_compute /** Descriptor for FFT scale kernels */ struct FFTScaleKernelInfo { - float scale{ 0.f }; /**< Axis to perform the kernel on. */ - bool conjugate{ true }; /**< Flag to conjugate the output/ */ + float scale{0.f}; /**< Axis to perform the kernel on. */ + bool conjugate{true}; /**< Flag to conjugate the output/ */ }; /** Descriptor for FFT digit reverse kernels */ struct FFTDigitReverseKernelInfo { - unsigned int axis{ 0 }; /**< Axis to perform the kernel on. */ - bool conjugate{ false }; /**< Flag to conjugate the output/ */ + unsigned int axis{0}; /**< Axis to perform the kernel on. */ + bool conjugate{false}; /**< Flag to conjugate the output/ */ }; /** Descriptor used by the FFT core kernels */ struct FFTRadixStageKernelInfo { - unsigned int axis{ 0 }; /**< Axis to run the kernel on. */ - unsigned int radix{ 0 }; /**< Radix to use. */ - unsigned int Nx{ 0 }; /**< Nx coefficient. */ - bool is_first_stage{ false }; /**< Flags if the FFT kernels is the first stage of a decomposed FFT. */ + unsigned int axis{0}; /**< Axis to run the kernel on. */ + unsigned int radix{0}; /**< Radix to use. */ + unsigned int Nx{0}; /**< Nx coefficient. */ + bool is_first_stage{false}; /**< Flags if the FFT kernels is the first stage of a decomposed FFT. */ }; class ITensorInfo; @@ -58,89 +58,102 @@ class ITensorInfo; struct GEMMKernelInfo { GEMMKernelInfo() = default; - GEMMKernelInfo( - unsigned int im, - unsigned int in, - unsigned int ik, - unsigned int idepth_output_gemm3d, - bool ireinterpret_input_as_3d, - bool ibroadcast_bias, - bool ifp_mixed_precision, - bool ihas_pad_y, - ActivationLayerInfo iactivation_info, - int inmult_transpose1xW_width, - int imult_interleave4x4_height, - GEMMLHSMatrixInfo ilhs_info, - GEMMRHSMatrixInfo irhs_info, - int32_t ina_offset, - int32_t inb_offset) - : m(im), n(in), k(ik), depth_output_gemm3d(idepth_output_gemm3d), reinterpret_input_as_3d(ireinterpret_input_as_3d), broadcast_bias(ibroadcast_bias), fp_mixed_precision(ifp_mixed_precision), - has_pad_y(ihas_pad_y), activation_info(iactivation_info), mult_transpose1xW_width(inmult_transpose1xW_width), mult_interleave4x4_height(imult_interleave4x4_height), lhs_info(ilhs_info), - rhs_info(irhs_info), a_offset(ina_offset), b_offset(inb_offset) + GEMMKernelInfo(unsigned int im, + unsigned int in, + unsigned int ik, + unsigned int idepth_output_gemm3d, + bool ireinterpret_input_as_3d, + bool ibroadcast_bias, + bool ifp_mixed_precision, + bool ihas_pad_y, + ActivationLayerInfo iactivation_info, + int inmult_transpose1xW_width, + int imult_interleave4x4_height, + GEMMLHSMatrixInfo ilhs_info, + GEMMRHSMatrixInfo irhs_info, + int32_t ina_offset, + int32_t inb_offset) + : m(im), + n(in), + k(ik), + depth_output_gemm3d(idepth_output_gemm3d), + reinterpret_input_as_3d(ireinterpret_input_as_3d), + broadcast_bias(ibroadcast_bias), + fp_mixed_precision(ifp_mixed_precision), + has_pad_y(ihas_pad_y), + activation_info(iactivation_info), + mult_transpose1xW_width(inmult_transpose1xW_width), + mult_interleave4x4_height(imult_interleave4x4_height), + lhs_info(ilhs_info), + rhs_info(irhs_info), + a_offset(ina_offset), + b_offset(inb_offset) { } - unsigned int m{ 0 }; /**< Number of LHS rows*/ - unsigned int n{ 0 }; /**< Number of RHS columns*/ - unsigned int k{ 0 }; /**< Number of LHS columns or RHS rows */ - unsigned int depth_output_gemm3d{ 0 }; /**< Depth of the output tensor in case is reinterpreted as 3D */ - bool reinterpret_input_as_3d{ false }; /**< Flag used to reinterpret the input as 3D */ - bool broadcast_bias{ false }; /**< Flag used to broadcast the bias addition */ - bool fp_mixed_precision{ false }; /**< Flag used to indicate wider accumulators (32 bit instead of 16 for FP16). */ - bool has_pad_y{ false }; /**< Flag used to indicate if the input/output tensors have internal pad on the y direction */ - ActivationLayerInfo activation_info{}; /**< Activation function to perform after the matrix multiplication */ - int mult_transpose1xW_width{ 1 }; /**< Multiplication factor for the width of the 1xW transposed block */ - int mult_interleave4x4_height{ 1 }; /**< Multiplication factor for the height of the 4x4 interleaved block */ - GEMMLHSMatrixInfo lhs_info{}; /**< LHS matrix information used to retrieve the number of rows processed by each thread */ - GEMMRHSMatrixInfo rhs_info{}; /**< RHS matrix information used for reshaping the RHS matrix */ - int32_t a_offset{ 0 }; /**< Offset to be added to each element of the matrix A */ - int32_t b_offset{ 0 }; /**< Offset to be added to each element of the matrix B */ - GEMMLowpOutputStageInfo output_stage{}; /**< GEMMLowp output stage information */ + unsigned int m{0}; /**< Number of LHS rows*/ + unsigned int n{0}; /**< Number of RHS columns*/ + unsigned int k{0}; /**< Number of LHS columns or RHS rows */ + unsigned int depth_output_gemm3d{0}; /**< Depth of the output tensor in case is reinterpreted as 3D */ + bool reinterpret_input_as_3d{false}; /**< Flag used to reinterpret the input as 3D */ + bool broadcast_bias{false}; /**< Flag used to broadcast the bias addition */ + bool fp_mixed_precision{false}; /**< Flag used to indicate wider accumulators (32 bit instead of 16 for FP16). */ + bool has_pad_y{ + false}; /**< Flag used to indicate if the input/output tensors have internal pad on the y direction */ + ActivationLayerInfo activation_info{}; /**< Activation function to perform after the matrix multiplication */ + int mult_transpose1xW_width{1}; /**< Multiplication factor for the width of the 1xW transposed block */ + int mult_interleave4x4_height{1}; /**< Multiplication factor for the height of the 4x4 interleaved block */ + GEMMLHSMatrixInfo + lhs_info{}; /**< LHS matrix information used to retrieve the number of rows processed by each thread */ + GEMMRHSMatrixInfo rhs_info{}; /**< RHS matrix information used for reshaping the RHS matrix */ + int32_t a_offset{0}; /**< Offset to be added to each element of the matrix A */ + int32_t b_offset{0}; /**< Offset to be added to each element of the matrix B */ + GEMMLowpOutputStageInfo output_stage{}; /**< GEMMLowp output stage information */ }; /** Compute descriptor used by the depthwise convolution native kernel */ struct DWCComputeKernelInfo { - unsigned int n0{ 1 }; /**< Number of columns processed by each thread */ - unsigned int m0{ 1 }; /**< Number of rows processed by each thread */ - bool export_input_to_cl_image{ false }; /**< Export input to cl_image */ - bool export_weights_to_cl_image{ false }; /**< Export the weights to cl_image */ + unsigned int n0{1}; /**< Number of columns processed by each thread */ + unsigned int m0{1}; /**< Number of rows processed by each thread */ + bool export_input_to_cl_image{false}; /**< Export input to cl_image */ + bool export_weights_to_cl_image{false}; /**< Export the weights to cl_image */ }; /** Compute descriptor used by the direct convolution kernel */ struct DirectConvComputeKernelInfo { - int32_t m0{ 1 }; /**< Number of rows to be processed by the kernel */ - int32_t n0{ 1 }; /**< Number of columns to be processed by the kernel */ - int32_t k0{ 1 }; /**< Number of partial accumulations to be processed in a single iteration by the kernel */ - bool export_weights_to_cl_image{ false }; /**< Flag to export the weights to cl_image */ - bool export_output_to_cl_image{ false }; /**< Flag to export the output to cl_image */ - bool export_input_to_cl_image{ false }; /**< Flag to export the input to cl_image */ + int32_t m0{1}; /**< Number of rows to be processed by the kernel */ + int32_t n0{1}; /**< Number of columns to be processed by the kernel */ + int32_t k0{1}; /**< Number of partial accumulations to be processed in a single iteration by the kernel */ + bool export_weights_to_cl_image{false}; /**< Flag to export the weights to cl_image */ + bool export_output_to_cl_image{false}; /**< Flag to export the output to cl_image */ + bool export_input_to_cl_image{false}; /**< Flag to export the input to cl_image */ }; /** Descriptor used by the softmax kernels */ struct SoftmaxKernelInfo { - float beta{ 1.f }; /**< A scaling factor for the exponent with default value 1.0 */ - bool is_log{ false }; /**< Flag used to perform Log Softmax operation */ - DataType input_data_type{ DataType::UNKNOWN }; /**< Input tensor data type */ - int32_t axis{ 0 }; /**< The dimension in which to apply softmax. */ + float beta{1.f}; /**< A scaling factor for the exponent with default value 1.0 */ + bool is_log{false}; /**< Flag used to perform Log Softmax operation */ + DataType input_data_type{DataType::UNKNOWN}; /**< Input tensor data type */ + int32_t axis{0}; /**< The dimension in which to apply softmax. */ }; /** Descriptor used by the direct convolution layer output stage kernels */ struct DirectConvolutionLayerOutputStageKernelInfo { - int32_t result_fixedpoint_multiplier{ 0 }; /**< Result output stage multiplier used for quantizing */ - int32_t result_shift{ 0 }; /**< Result output stage shift used for quantizing */ - int32_t result_offset_after_shift{ 0 }; /**< Result offset used for quantizing */ - DataType output_data_type{ DataType::UNKNOWN }; /**< Output tensor data type to use if the output is not initialized */ + int32_t result_fixedpoint_multiplier{0}; /**< Result output stage multiplier used for quantizing */ + int32_t result_shift{0}; /**< Result output stage shift used for quantizing */ + int32_t result_offset_after_shift{0}; /**< Result offset used for quantizing */ + DataType output_data_type{ + DataType::UNKNOWN}; /**< Output tensor data type to use if the output is not initialized */ }; struct InstanceNormalizationLayerKernelInfo { /** Default constructor */ - InstanceNormalizationLayerKernelInfo() - : InstanceNormalizationLayerKernelInfo(1.f, 0.f, 1e-12, true) + InstanceNormalizationLayerKernelInfo() : InstanceNormalizationLayerKernelInfo(1.f, 0.f, 1e-12, true) { } /** Constructor @@ -177,10 +190,10 @@ struct GEMMLowpReductionKernelInfo { } - int32_t k{ 0 }; /**< Number of matrix columns/rows */ - bool is_reshaped{ false }; /**< True if the input tensor has been reshaped */ - int32_t scalar{ 0 }; /**< Scalar value to multiply each reduced column/row by */ - bool mul_by_scalar{ false }; /**< True if each column/row reduction has to be multiplied by a scalar value */ + int32_t k{0}; /**< Number of matrix columns/rows */ + bool is_reshaped{false}; /**< True if the input tensor has been reshaped */ + int32_t scalar{0}; /**< Scalar value to multiply each reduced column/row by */ + bool mul_by_scalar{false}; /**< True if each column/row reduction has to be multiplied by a scalar value */ }; struct ScaleKernelInfo @@ -202,13 +215,13 @@ struct ScaleKernelInfo bool use_padding = true, bool align_corners = false, DataLayout data_layout = DataLayout::UNKNOWN) noexcept - : interpolation_policy{ interpolation_policy }, - border_mode{ border_mode }, - constant_border_value{ constant_border_value }, - sampling_policy{ sampling_policy }, - use_padding{ use_padding }, - align_corners{ align_corners }, - data_layout{ data_layout } + : interpolation_policy{interpolation_policy}, + border_mode{border_mode}, + constant_border_value{constant_border_value}, + sampling_policy{sampling_policy}, + use_padding{use_padding}, + align_corners{align_corners}, + data_layout{data_layout} { } @@ -224,16 +237,17 @@ struct ScaleKernelInfo struct MatMulKernelInfo { MatMulKernelInfo() = default; - MatMulKernelInfo(bool adj_lhs, bool adj_rhs, int m0 = 1, int n0 = 1, int k0 = 1, bool export_rhs_to_cl_image = false) - : adj_lhs{ adj_lhs }, adj_rhs{ adj_rhs }, m0{ m0 }, n0{ n0 }, k0{ k0 }, export_rhs_to_cl_image{ export_rhs_to_cl_image } + MatMulKernelInfo( + bool adj_lhs, bool adj_rhs, int m0 = 1, int n0 = 1, int k0 = 1, bool export_rhs_to_cl_image = false) + : adj_lhs{adj_lhs}, adj_rhs{adj_rhs}, m0{m0}, n0{n0}, k0{k0}, export_rhs_to_cl_image{export_rhs_to_cl_image} { } - bool adj_lhs{ false }; /**< Get Adjoint LHS flag value */ - bool adj_rhs{ false }; /**< Get Adjoint RHS flag value */ - int m0{ 1 }; /**< Number of output rows processed by each work-item*/ - int n0{ 1 }; /**< Number of output columns processed by each work-item*/ - int k0{ 1 }; /**< Number of inner accumulations */ - bool export_rhs_to_cl_image{ false }; /**< Flag to know whether the RHS tensor should be exported to cl_image*/ + bool adj_lhs{false}; /**< Get Adjoint LHS flag value */ + bool adj_rhs{false}; /**< Get Adjoint RHS flag value */ + int m0{1}; /**< Number of output rows processed by each work-item*/ + int n0{1}; /**< Number of output columns processed by each work-item*/ + int k0{1}; /**< Number of inner accumulations */ + bool export_rhs_to_cl_image{false}; /**< Flag to know whether the RHS tensor should be exported to cl_image*/ }; } // namespace arm_compute #endif // ACL_ARM_COMPUTE_CORE_KERNELDESCRIPTORS_H diff --git a/arm_compute/core/Log.h b/arm_compute/core/Log.h index bc0ecb802e..03b861f765 100644 --- a/arm_compute/core/Log.h +++ b/arm_compute/core/Log.h @@ -34,11 +34,11 @@ #define ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER() \ do \ { \ - if(arm_compute::logging::LoggerRegistry::get().logger("CORE") == nullptr) \ + if (arm_compute::logging::LoggerRegistry::get().logger("CORE") == nullptr) \ { \ arm_compute::logging::LoggerRegistry::get().create_reserved_loggers(); \ } \ - } while(false) + } while (false) #else /* ARM_COMPUTE_LOGGING_ENABLED */ #define ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER() #endif /* ARM_COMPUTE_LOGGING_ENABLED */ @@ -53,7 +53,7 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_MSG("CORE", log_level, msg); \ - } while(false) + } while (false) /** Log a message with format to the core system logger * @@ -66,7 +66,7 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_MSG_WITH_FORMAT("CORE", log_level, fmt, __VA_ARGS__); \ - } while(false) + } while (false) /** Log a stream to the core system logger * @@ -78,7 +78,7 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_STREAM("CORE", log_level, ss); \ - } while(false) + } while (false) /** Log information level message to the core system logger * @@ -89,7 +89,7 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_MSG_CORE(arm_compute::logging::LogLevel::INFO, msg); \ - } while(false) + } while (false) /** Log information level formatted message to the core system logger * @@ -101,7 +101,7 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_MSG_WITH_FORMAT_CORE(arm_compute::logging::LogLevel::INFO, #fmt, __VA_ARGS__); \ - } while(false) + } while (false) /** Log information level stream to the core system logger * @@ -112,6 +112,6 @@ { \ ARM_COMPUTE_CREATE_DEFAULT_CORE_LOGGER(); \ ARM_COMPUTE_LOG_STREAM_CORE(arm_compute::logging::LogLevel::INFO, ss); \ - } while(false) + } while (false) #endif /* ARM_COMPUTE_LOGGING_MACROS_H */ diff --git a/arm_compute/core/PixelValue.h b/arm_compute/core/PixelValue.h index 790f58a793..0b4df4f2e2 100644 --- a/arm_compute/core/PixelValue.h +++ b/arm_compute/core/PixelValue.h @@ -24,8 +24,8 @@ #ifndef ARM_COMPUTE_PIXELVALUE_H #define ARM_COMPUTE_PIXELVALUE_H -#include "arm_compute/core/Types.h" #include "arm_compute/core/QuantizationInfo.h" +#include "arm_compute/core/Types.h" #include <cstdint> @@ -36,11 +36,7 @@ class PixelValue { public: /** Default constructor: value initialized to 0 */ - PixelValue() noexcept - : value - { - int64_t(0) - } + PixelValue() noexcept : value{int64_t(0)} { } /** Initialize the union with a pixel value of chosen datatype @@ -49,10 +45,9 @@ public: * @param[in] datatype DataType that @p v have to be stored * @param[in] qinfo (Optional) QuantizationInfo to apply in case of quantized data types to @p v */ - PixelValue(double v, DataType datatype, QuantizationInfo qinfo = QuantizationInfo()) - : PixelValue() + PixelValue(double v, DataType datatype, QuantizationInfo qinfo = QuantizationInfo()) : PixelValue() { - switch(datatype) + switch (datatype) { case DataType::U8: value.u8 = static_cast<uint8_t>(v); @@ -112,8 +107,7 @@ public: * * @param[in] v S8 value. */ - PixelValue(int8_t v) - : PixelValue() + PixelValue(int8_t v) : PixelValue() { value.s8 = v; } @@ -121,8 +115,7 @@ public: * * @param[in] v U8 value. */ - PixelValue(uint8_t v) - : PixelValue() + PixelValue(uint8_t v) : PixelValue() { value.u8 = v; } @@ -130,8 +123,7 @@ public: * * @param[in] v U16 value. */ - PixelValue(uint16_t v) - : PixelValue() + PixelValue(uint16_t v) : PixelValue() { value.u16 = v; } @@ -139,8 +131,7 @@ public: * * @param[in] v S16 value. */ - PixelValue(int16_t v) - : PixelValue() + PixelValue(int16_t v) : PixelValue() { value.s16 = v; } @@ -148,8 +139,7 @@ public: * * @param[in] v U32 value. */ - PixelValue(uint32_t v) - : PixelValue() + PixelValue(uint32_t v) : PixelValue() { value.u32 = v; } @@ -157,8 +147,7 @@ public: * * @param[in] v S32 value. */ - PixelValue(int32_t v) - : PixelValue() + PixelValue(int32_t v) : PixelValue() { value.s32 = v; } @@ -167,8 +156,7 @@ public: * * @param[in] v U64 value. */ - PixelValue(uint64_t v) - : PixelValue() + PixelValue(uint64_t v) : PixelValue() { value.u64 = v; } @@ -176,8 +164,7 @@ public: * * @param[in] v S64 value. */ - PixelValue(int64_t v) - : PixelValue() + PixelValue(int64_t v) : PixelValue() { value.s64 = v; } @@ -185,8 +172,7 @@ public: * * @param[in] v F16 value. */ - PixelValue(bfloat16 v) - : PixelValue() + PixelValue(bfloat16 v) : PixelValue() { value.bf16 = v; } @@ -194,8 +180,7 @@ public: * * @param[in] v F16 value. */ - PixelValue(half v) - : PixelValue() + PixelValue(half v) : PixelValue() { value.f16 = v; } @@ -203,8 +188,7 @@ public: * * @param[in] v F32 value. */ - PixelValue(float v) - : PixelValue() + PixelValue(float v) : PixelValue() { value.f32 = v; } @@ -212,8 +196,7 @@ public: * * @param[in] v F64 value. */ - PixelValue(double v) - : PixelValue() + PixelValue(double v) : PixelValue() { value.f64 = v; } @@ -221,23 +204,23 @@ public: * Use the field corresponding to the image format */ union - { - uint64_t u64; /**< Single channel U64 */ - int64_t s64; /**< Single channel S64 */ - uint8_t rgb[3]; /**< 3 channels: RGB888 */ - uint8_t yuv[3]; /**< 3 channels: Any YUV format */ - uint8_t rgbx[4]; /**< 4 channels: RGBX8888 */ - double f64; /**< Single channel double */ - float f32; /**< Single channel float 32 */ - half f16; /**< Single channel F16 */ - bfloat16 bf16; /**< Single channel brain floating-point number */ - uint8_t u8; /**< Single channel U8 */ - int8_t s8; /**< Single channel S8 */ - uint16_t u16; /**< Single channel U16 */ - int16_t s16; /**< Single channel S16 */ - uint32_t u32; /**< Single channel U32 */ - int32_t s32; /**< Single channel S32 */ - } value; + { + uint64_t u64; /**< Single channel U64 */ + int64_t s64; /**< Single channel S64 */ + uint8_t rgb[3]; /**< 3 channels: RGB888 */ + uint8_t yuv[3]; /**< 3 channels: Any YUV format */ + uint8_t rgbx[4]; /**< 4 channels: RGBX8888 */ + double f64; /**< Single channel double */ + float f32; /**< Single channel float 32 */ + half f16; /**< Single channel F16 */ + bfloat16 bf16; /**< Single channel brain floating-point number */ + uint8_t u8; /**< Single channel U8 */ + int8_t s8; /**< Single channel S8 */ + uint16_t u16; /**< Single channel U16 */ + int16_t s16; /**< Single channel S16 */ + uint32_t u32; /**< Single channel U32 */ + int32_t s32; /**< Single channel S32 */ + } value; /** Interpret the pixel value as a U8 * * @param[out] v Returned value diff --git a/arm_compute/core/QuantizationInfo.h b/arm_compute/core/QuantizationInfo.h index 8fa513eee1..471b8c57ab 100644 --- a/arm_compute/core/QuantizationInfo.h +++ b/arm_compute/core/QuantizationInfo.h @@ -26,6 +26,7 @@ #include "arm_compute/core/Rounding.h" #include "arm_compute/core/utils/misc/Utility.h" + #include "support/ToolchainSupport.h" #include <vector> @@ -41,8 +42,7 @@ using qasymm16_t = uint16_t; /**< 16 bit quantized asymmetric scalar value struct UniformQuantizationInfo { /** Default constructor */ - UniformQuantizationInfo() - : scale(0.f), offset(0) + UniformQuantizationInfo() : scale(0.f), offset(0) { } /** Constructor @@ -50,8 +50,7 @@ struct UniformQuantizationInfo * @param[in] scale Quantization scale * @param[in] offset Quantization offset */ - UniformQuantizationInfo(float scale, int32_t offset) - : scale(scale), offset(offset) + UniformQuantizationInfo(float scale, int32_t offset) : scale(scale), offset(offset) { } /** Checks if the scale and offset are both zero */ @@ -69,9 +68,7 @@ class QuantizationInfo { public: /** Default constructor */ - QuantizationInfo() noexcept - : _scale(), - _offset() + QuantizationInfo() noexcept : _scale(), _offset() { } /** Construct quantization info. @@ -80,8 +77,7 @@ public: * * @param[in] scale Scale. */ - QuantizationInfo(float scale) - : _scale(1, scale), _offset() + QuantizationInfo(float scale) : _scale(1, scale), _offset() { } /** Construct quantization info. @@ -91,8 +87,7 @@ public: * @param[in] scale Scale. * @param[in] offset Offset. */ - QuantizationInfo(float scale, int offset) - : _scale(1, scale), _offset(1, offset) + QuantizationInfo(float scale, int offset) : _scale(1, scale), _offset(1, offset) { } /** Construct quantization info. @@ -101,8 +96,7 @@ public: * * @param[in] scale Scale. */ - QuantizationInfo(std::vector<float> scale) - : _scale(scale), _offset() + QuantizationInfo(std::vector<float> scale) : _scale(scale), _offset() { } /** Construct quantization info. @@ -112,8 +106,7 @@ public: * @param[in] scale Scale. * @param[in] offset Offset. */ - QuantizationInfo(std::vector<float> scale, std::vector<int32_t> offset) - : _scale(scale), _offset(offset) + QuantizationInfo(std::vector<float> scale, std::vector<int32_t> offset) : _scale(scale), _offset(offset) { } /** Scale vector accessor @@ -208,8 +201,7 @@ inline bool operator!=(const UniformQuantizationInfo &lhs, const UniformQuantiza template <typename QUANTIZED_TYPE = uint8_t> struct Qasymm8QuantizationHelper { - static_assert(std::is_same<QUANTIZED_TYPE, uint8_t>::value - || std::is_same<QUANTIZED_TYPE, int8_t>::value, + static_assert(std::is_same<QUANTIZED_TYPE, uint8_t>::value || std::is_same<QUANTIZED_TYPE, int8_t>::value, "quantized type should be either uint8_t or int8_t."); /** Quantize a value given a 8-bit asymmetric quantization scheme @@ -234,9 +226,10 @@ struct Qasymm8QuantizationHelper * * @return Quantized value */ - static inline QUANTIZED_TYPE quantize(float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy) + static inline QUANTIZED_TYPE + quantize(float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy) { - if(rounding_policy == RoundingPolicy::TO_NEAREST_UP) + if (rounding_policy == RoundingPolicy::TO_NEAREST_UP) { return quantize(value, qinfo); } @@ -254,7 +247,8 @@ struct Qasymm8QuantizationHelper * * @return Quantized value */ - static inline QUANTIZED_TYPE quantize(float value, const QuantizationInfo &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) + static inline QUANTIZED_TYPE + quantize(float value, const QuantizationInfo &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) { const UniformQuantizationInfo uqinfo = qinfo.uniform(); ARM_COMPUTE_ERROR_ON(uqinfo.scale == 0); @@ -297,7 +291,8 @@ struct Qasymm8QuantizationHelper * @return Quantized value */ template <typename INFO_TYPE> -inline uint8_t quantize_qasymm8(float value, const INFO_TYPE &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) +inline uint8_t +quantize_qasymm8(float value, const INFO_TYPE &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) { return Qasymm8QuantizationHelper<uint8_t>::quantize(value, qinfo, rounding_policy); } @@ -311,7 +306,9 @@ inline uint8_t quantize_qasymm8(float value, const INFO_TYPE &qinfo, RoundingPol * @return Quantized value */ template <typename INFO_TYPE> -inline int8_t quantize_qasymm8_signed(float value, const INFO_TYPE &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) +inline int8_t quantize_qasymm8_signed(float value, + const INFO_TYPE &qinfo, + RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) { return Qasymm8QuantizationHelper<int8_t>::quantize(value, qinfo, rounding_policy); } @@ -441,7 +438,9 @@ inline float dequantize(uint16_t value, float scale, int32_t offset) * * @return Quantized value */ -inline int16_t quantize_qsymm16(float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) +inline int16_t quantize_qsymm16(float value, + const UniformQuantizationInfo &qinfo, + RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) { int quantized = arm_compute::round(value / qinfo.scale, rounding_policy); quantized = arm_compute::utility::clamp<int, int16_t>(quantized); @@ -492,7 +491,9 @@ inline float dequantize_qsymm16(int16_t value, const QuantizationInfo &qinfo) * * @return Quantized value */ -inline uint16_t quantize_qasymm16(float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) +inline uint16_t quantize_qasymm16(float value, + const UniformQuantizationInfo &qinfo, + RoundingPolicy rounding_policy = RoundingPolicy::TO_NEAREST_UP) { int quantized = arm_compute::round(value / qinfo.scale, rounding_policy) + qinfo.offset; quantized = arm_compute::utility::clamp<int, uint16_t>(quantized); @@ -565,7 +566,8 @@ inline float dequantize_qasymm16(uint16_t value, const QuantizationInfo &qinfo) * z_n = - z_i * s_i / s_o + z_o * */ -inline UniformQuantizationInfo compute_requantization_scale_offset(const UniformQuantizationInfo &uqinfo_in, const UniformQuantizationInfo &uqinfo_out) +inline UniformQuantizationInfo compute_requantization_scale_offset(const UniformQuantizationInfo &uqinfo_in, + const UniformQuantizationInfo &uqinfo_out) { float scale_to_apply = uqinfo_out.scale; int32_t offset_to_apply = uqinfo_out.offset; diff --git a/arm_compute/core/Rounding.h b/arm_compute/core/Rounding.h index b6817b5107..30a5a0fe9d 100644 --- a/arm_compute/core/Rounding.h +++ b/arm_compute/core/Rounding.h @@ -42,5 +42,5 @@ enum class RoundingPolicy * @return Rounded value of the argument x. */ int round(float x, RoundingPolicy rounding_policy); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_ROUNDING_H */ diff --git a/arm_compute/core/Size2D.h b/arm_compute/core/Size2D.h index f3e9bea4c7..672b392050 100644 --- a/arm_compute/core/Size2D.h +++ b/arm_compute/core/Size2D.h @@ -41,9 +41,7 @@ public: * @param[in] w Width of the image or rectangle * @param[in] h Height of the image or rectangle */ - Size2D(size_t w, size_t h) noexcept - : width(w), - height(h) + Size2D(size_t w, size_t h) noexcept : width(w), height(h) { } /** The area of the image or rectangle calculated as (width * height) @@ -90,5 +88,5 @@ public: size_t width = {}; /**< Width of the image region or rectangle */ size_t height = {}; /**< Height of the image region or rectangle */ }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_SIZE2D_H */ diff --git a/arm_compute/core/Size3D.h b/arm_compute/core/Size3D.h index 4241ed4f7e..e2dc6fe012 100644 --- a/arm_compute/core/Size3D.h +++ b/arm_compute/core/Size3D.h @@ -40,8 +40,7 @@ public: * @param[in] h Height of the 3D shape or object * @param[in] d Depth of the 3D shape or object */ - Size3D(size_t w, size_t h, size_t d) noexcept - : width(w), height(h), depth(d) + Size3D(size_t w, size_t h, size_t d) noexcept : width(w), height(h), depth(d) { } diff --git a/arm_compute/core/Steps.h b/arm_compute/core/Steps.h index 208fc4b294..6b261becc0 100644 --- a/arm_compute/core/Steps.h +++ b/arm_compute/core/Steps.h @@ -45,8 +45,7 @@ public: * @param[in] steps Values to initialize the steps. */ template <typename... Ts> - Steps(Ts... steps) - : Dimensions{ steps... } + Steps(Ts... steps) : Dimensions{steps...} { // Initialize empty dimensions to 1 std::fill(_id.begin() + _num_dimensions, _id.end(), 1); @@ -62,5 +61,5 @@ public: /** Default destructor */ ~Steps() = default; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_STEPS_H*/ diff --git a/arm_compute/core/Strides.h b/arm_compute/core/Strides.h index b582d066f7..627b219987 100644 --- a/arm_compute/core/Strides.h +++ b/arm_compute/core/Strides.h @@ -43,8 +43,7 @@ public: * @param[in] strides Values to initialize the strides. */ template <typename... Ts> - constexpr Strides(Ts... strides) - : Dimensions{ strides... } + constexpr Strides(Ts... strides) : Dimensions{strides...} { } /** Allow instances of this class to be copy constructed */ diff --git a/arm_compute/core/SubTensorInfo.h b/arm_compute/core/SubTensorInfo.h index 21703b0d93..7a3ee2cfd0 100644 --- a/arm_compute/core/SubTensorInfo.h +++ b/arm_compute/core/SubTensorInfo.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_SUBTENSORINFO_H #define ARM_COMPUTE_SUBTENSORINFO_H -#include "arm_compute/core/ITensorInfo.h" - #include "arm_compute/core/Coordinates.h" #include "arm_compute/core/Helpers.h" +#include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/Strides.h" #include "arm_compute/core/TensorInfo.h" #include "arm_compute/core/TensorShape.h" @@ -73,7 +72,7 @@ public: // Inherited methods overridden: std::unique_ptr<ITensorInfo> clone() const override; - ITensorInfo &set_data_type(DataType data_type) override + ITensorInfo &set_data_type(DataType data_type) override { ARM_COMPUTE_ERROR_ON(_parent == nullptr); _parent->set_data_type(data_type); @@ -143,7 +142,7 @@ public: return _parent->offset_element_in_bytes(_coords); } int32_t offset_element_in_bytes(const Coordinates &pos) const override; - size_t element_size() const override + size_t element_size() const override { ARM_COMPUTE_ERROR_ON(_parent == nullptr); return _parent->element_size(); @@ -227,7 +226,7 @@ public: { ARM_COMPUTE_ERROR_ON(_parent == nullptr); // Check if subtensor is valid if parent is configured - if(_parent->tensor_shape().total_size() != 0) + if (_parent->tensor_shape().total_size() != 0) { ARM_COMPUTE_ERROR_ON_INVALID_SUBTENSOR_VALID_REGION(_parent->valid_region(), valid_region); } diff --git a/arm_compute/core/TensorInfo.h b/arm_compute/core/TensorInfo.h index e738a797b2..b18f750427 100644 --- a/arm_compute/core/TensorInfo.h +++ b/arm_compute/core/TensorInfo.h @@ -24,15 +24,14 @@ #ifndef ARM_COMPUTE_TENSORINFO_H #define ARM_COMPUTE_TENSORINFO_H -#include "arm_compute/core/ITensorInfo.h" - -#include "ITensorInfo.h" #include "arm_compute/core/Coordinates.h" #include "arm_compute/core/Helpers.h" +#include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/Strides.h" #include "arm_compute/core/TensorShape.h" #include "arm_compute/core/Types.h" +#include "ITensorInfo.h" #include <cstddef> #include <memory> @@ -112,7 +111,10 @@ public: * @param[in] data_type Data type to use for each tensor element * @param[in] quantization_info The quantization settings for the tensor data. */ - TensorInfo(const TensorShape &tensor_shape, size_t num_channels, DataType data_type, QuantizationInfo quantization_info); + TensorInfo(const TensorShape &tensor_shape, + size_t num_channels, + DataType data_type, + QuantizationInfo quantization_info); /** Initialize the tensor info with just a format. * @@ -136,7 +138,11 @@ public: * @param[in] offset_first_element_in_bytes Offset in bytes from the beginning of memory allocation to access the first element. * @param[in] total_size_in_bytes Size in bytes of the memory allocation (including the offset to the first element). */ - void init(const TensorShape &tensor_shape, Format format, const Strides &strides_in_bytes, size_t offset_first_element_in_bytes, size_t total_size_in_bytes); + void init(const TensorShape &tensor_shape, + Format format, + const Strides &strides_in_bytes, + size_t offset_first_element_in_bytes, + size_t total_size_in_bytes); /** Initialize the tensor info with just a format. * @@ -164,8 +170,12 @@ public: * @param[in] offset_first_element_in_bytes Offset in bytes from the beginning of memory allocation to access the first element. * @param[in] total_size_in_bytes Size in bytes of the memory allocation (including the offset to the first element). */ - void init(const TensorShape &tensor_shape, size_t num_channels, DataType data_type, const Strides &strides_in_bytes, size_t offset_first_element_in_bytes, - size_t total_size_in_bytes); + void init(const TensorShape &tensor_shape, + size_t num_channels, + DataType data_type, + const Strides &strides_in_bytes, + size_t offset_first_element_in_bytes, + size_t total_size_in_bytes); /** Initialize the metadata structure for the given tensor shape and single-plane format, (Padding is automatically calculated) * * @note The padding used by this method is really conservative so that the tensor can be used for most functions. @@ -191,19 +201,19 @@ public: // Inherited methods overridden: std::unique_ptr<ITensorInfo> clone() const override; - ITensorInfo &set_data_type(DataType data_type) override; - ITensorInfo &set_num_channels(int num_channels) override; - ITensorInfo &set_format(Format format) override; - ITensorInfo &set_tensor_shape(const TensorShape &shape) override; - ITensorInfo &set_tensor_dims_state(const TensorDimsState &state) override; - ITensorInfo &set_quantization_info(const QuantizationInfo &quantization_info) override; - ITensorInfo &set_data_layout(const DataLayout &data_layout) override; - ITensorInfo &reset_padding() override; - bool auto_padding() override; - ITensorInfo &set_lock_paddings(bool flag) override; - bool lock_paddings() const override; - bool extend_padding(const PaddingSize &padding) override; - size_t dimension(size_t index) const override + ITensorInfo &set_data_type(DataType data_type) override; + ITensorInfo &set_num_channels(int num_channels) override; + ITensorInfo &set_format(Format format) override; + ITensorInfo &set_tensor_shape(const TensorShape &shape) override; + ITensorInfo &set_tensor_dims_state(const TensorDimsState &state) override; + ITensorInfo &set_quantization_info(const QuantizationInfo &quantization_info) override; + ITensorInfo &set_data_layout(const DataLayout &data_layout) override; + ITensorInfo &reset_padding() override; + bool auto_padding() override; + ITensorInfo &set_lock_paddings(bool flag) override; + bool lock_paddings() const override; + bool extend_padding(const PaddingSize &padding) override; + size_t dimension(size_t index) const override { return _tensor_shape[index]; } @@ -220,7 +230,7 @@ public: return _offset_first_element_in_bytes; } int32_t offset_element_in_bytes(const Coordinates &pos) const override; - size_t element_size() const override + size_t element_size() const override { return data_size_from_type(_data_type) * _num_channels; } @@ -266,7 +276,8 @@ public: } bool is_dynamic() const override { - return std::find(std::cbegin(_dims_state), std::cend(_dims_state), get_dynamic_state_value()) != std::cend(_dims_state); + return std::find(std::cbegin(_dims_state), std::cend(_dims_state), get_dynamic_state_value()) != + std::cend(_dims_state); } bool are_values_constant() const override { @@ -343,11 +354,15 @@ private: */ inline bool operator==(const TensorInfo &lhs, const TensorInfo &rhs) { - return (lhs._total_size == rhs._total_size) && (lhs._offset_first_element_in_bytes == rhs._offset_first_element_in_bytes) && (lhs._strides_in_bytes == rhs._strides_in_bytes) - && (lhs._num_channels == rhs._num_channels) && (lhs._tensor_shape == rhs._tensor_shape) && (lhs._dims_state == rhs._dims_state) && (lhs._data_type == rhs._data_type) && (lhs._format == rhs._format) - && (lhs._is_resizable == rhs._is_resizable) && (lhs._valid_region == rhs._valid_region) && (lhs._padding == rhs._padding) && (lhs._quantization_info == rhs._quantization_info) - && (lhs._data_layout == rhs._data_layout) && (lhs._are_values_constant == rhs._are_values_constant) - && (lhs._id == rhs._id); + return (lhs._total_size == rhs._total_size) && + (lhs._offset_first_element_in_bytes == rhs._offset_first_element_in_bytes) && + (lhs._strides_in_bytes == rhs._strides_in_bytes) && (lhs._num_channels == rhs._num_channels) && + (lhs._tensor_shape == rhs._tensor_shape) && (lhs._dims_state == rhs._dims_state) && + (lhs._data_type == rhs._data_type) && (lhs._format == rhs._format) && + (lhs._is_resizable == rhs._is_resizable) && (lhs._valid_region == rhs._valid_region) && + (lhs._padding == rhs._padding) && (lhs._quantization_info == rhs._quantization_info) && + (lhs._data_layout == rhs._data_layout) && (lhs._are_values_constant == rhs._are_values_constant) && + (lhs._id == rhs._id); } } // namespace arm_compute #endif /*ARM_COMPUTE_TENSORINFO_H */ diff --git a/arm_compute/core/TensorShape.h b/arm_compute/core/TensorShape.h index 4c9186ac64..c1707e262f 100644 --- a/arm_compute/core/TensorShape.h +++ b/arm_compute/core/TensorShape.h @@ -44,11 +44,10 @@ public: * @param[in] dims Values to initialize the dimensions. */ template <typename... Ts> - TensorShape(Ts... dims) - : Dimensions{ dims... } + TensorShape(Ts... dims) : Dimensions{dims...} { // Initialize unspecified dimensions to 1 - if(_num_dimensions > 0) + if (_num_dimensions > 0) { std::fill(_id.begin() + _num_dimensions, _id.end(), 1); } @@ -79,7 +78,7 @@ public: TensorShape &set(size_t dimension, size_t value, bool apply_dim_correction = true, bool increase_dim_unit = true) { // Clear entire shape if one dimension is zero - if(value == 0) + if (value == 0) { _num_dimensions = 0; std::fill(_id.begin(), _id.end(), 0); @@ -94,7 +93,7 @@ public: Dimensions::set(dimension, value, increase_dim_unit); // Correct number dimensions to ignore trailing dimensions of size 1 - if(apply_dim_correction) + if (apply_dim_correction) { apply_dimension_correction(); } @@ -123,7 +122,7 @@ public: std::fill(_id.begin() + _num_dimensions, _id.end(), 1); // Correct number dimensions to ignore trailing dimensions of size 1 - if(apply_dim_correction) + if (apply_dim_correction) { apply_dimension_correction(); } @@ -212,26 +211,26 @@ public: * @return The broadcasted shape or an empty shape if the shapes are not broadcast compatible. */ template <typename... Shapes> - static TensorShape broadcast_shape(const Shapes &... shapes) + static TensorShape broadcast_shape(const Shapes &...shapes) { TensorShape bc_shape; - auto broadcast = [&bc_shape](const TensorShape & other) + auto broadcast = [&bc_shape](const TensorShape &other) { - if(bc_shape.num_dimensions() == 0) + if (bc_shape.num_dimensions() == 0) { bc_shape = other; } - else if(other.num_dimensions() != 0) + else if (other.num_dimensions() != 0) { - for(size_t d = 0; d < TensorShape::num_max_dimensions; ++d) + for (size_t d = 0; d < TensorShape::num_max_dimensions; ++d) { const size_t dim_min = std::min(bc_shape[d], other[d]); const size_t dim_max = std::max(bc_shape[d], other[d]); - if((dim_min != 1) && (dim_min != dim_max)) + if ((dim_min != 1) && (dim_min != dim_max)) { - bc_shape = TensorShape{ 0U }; + bc_shape = TensorShape{0U}; break; } @@ -249,9 +248,9 @@ private: /** Remove trailing dimensions of size 1 from the reported number of dimensions. */ void apply_dimension_correction() { - for(int i = static_cast<int>(_num_dimensions) - 1; i > 0; --i) + for (int i = static_cast<int>(_num_dimensions) - 1; i > 0; --i) { - if(_id[i] == 1) + if (_id[i] == 1) { --_num_dimensions; } @@ -262,5 +261,5 @@ private: } } }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_TENSORSHAPE_H*/ diff --git a/arm_compute/core/Types.h b/arm_compute/core/Types.h index 9264cefe3e..6b51af17d4 100644 --- a/arm_compute/core/Types.h +++ b/arm_compute/core/Types.h @@ -59,13 +59,13 @@ /** The following symbols have been moved to: * MatMulInfo */ -#include "arm_compute/function_info/MatMulInfo.h" - #include "arm_compute/core/Coordinates.h" #include "arm_compute/core/Size2D.h" #include "arm_compute/core/Size3D.h" #include "arm_compute/core/TensorShape.h" #include "arm_compute/core/utils/misc/Macros.h" +#include "arm_compute/function_info/MatMulInfo.h" + #include "support/Bfloat16.h" #include <cmath> @@ -143,8 +143,7 @@ enum class ComparisonOperation struct ValidRegion { /** Default constructor */ - ValidRegion() - : anchor{}, shape{} + ValidRegion() : anchor{}, shape{} { } @@ -165,8 +164,7 @@ struct ValidRegion * @param[in] a_shape Shape of the valid region. * */ - ValidRegion(const Coordinates &an_anchor, const TensorShape &a_shape) - : anchor{ an_anchor }, shape{ a_shape } + ValidRegion(const Coordinates &an_anchor, const TensorShape &a_shape) : anchor{an_anchor}, shape{a_shape} { anchor.set_num_dimensions(std::max(anchor.num_dimensions(), shape.num_dimensions())); } @@ -179,7 +177,7 @@ struct ValidRegion * */ ValidRegion(const Coordinates &an_anchor, const TensorShape &a_shape, size_t num_dimensions) - : anchor{ an_anchor }, shape{ a_shape } + : anchor{an_anchor}, shape{a_shape} { ARM_COMPUTE_ERROR_ON(num_dimensions < std::max(anchor.num_dimensions(), shape.num_dimensions())); anchor.set_num_dimensions(num_dimensions); @@ -241,32 +239,24 @@ enum class BorderMode struct BorderSize { /** Empty border, i.e. no border */ - constexpr BorderSize() noexcept - : top{ 0 }, - right{ 0 }, - bottom{ 0 }, - left{ 0 } + constexpr BorderSize() noexcept : top{0}, right{0}, bottom{0}, left{0} { } /** Border with equal size around the 2D plane */ - explicit constexpr BorderSize(unsigned int size) noexcept - : top{ size }, - right{ size }, - bottom{ size }, - left{ size } + explicit constexpr BorderSize(unsigned int size) noexcept : top{size}, right{size}, bottom{size}, left{size} { } /** Border with same size for top/bottom and left/right */ constexpr BorderSize(unsigned int top_bottom, unsigned int left_right) - : top{ top_bottom }, right{ left_right }, bottom{ top_bottom }, left{ left_right } + : top{top_bottom}, right{left_right}, bottom{top_bottom}, left{left_right} { } /** Border with different sizes */ constexpr BorderSize(unsigned int top, unsigned int right, unsigned int bottom, unsigned int left) - : top{ top }, right{ right }, bottom{ bottom }, left{ left } + : top{top}, right{right}, bottom{bottom}, left{left} { } @@ -371,7 +361,7 @@ enum class InterpolationPolicy { NEAREST_NEIGHBOR, /**< Output values are defined to match the source pixel whose center is nearest to the sample position */ BILINEAR, /**< Output values are defined by bilinear interpolation between the pixels */ - AREA, /**< Output values are determined by averaging the source pixels whose areas fall under the area of the destination pixel, projected onto the source image */ + AREA, /**< Output values are determined by averaging the source pixels whose areas fall under the area of the destination pixel, projected onto the source image */ }; /** Bilinear Interpolation method used by LKTracker */ @@ -478,12 +468,12 @@ enum class NormType */ struct DetectionWindow { - uint16_t x{ 0 }; /**< Top-left x coordinate */ - uint16_t y{ 0 }; /**< Top-left y coordinate */ - uint16_t width{ 0 }; /**< Width of the detection window */ - uint16_t height{ 0 }; /**< Height of the detection window */ - uint16_t idx_class{ 0 }; /**< Index of the class */ - float score{ 0.f }; /**< Confidence value for the detection window */ + uint16_t x{0}; /**< Top-left x coordinate */ + uint16_t y{0}; /**< Top-left y coordinate */ + uint16_t width{0}; /**< Width of the detection window */ + uint16_t height{0}; /**< Height of the detection window */ + uint16_t idx_class{0}; /**< Index of the class */ + float score{0.f}; /**< Confidence value for the detection window */ }; /** Available pooling types */ @@ -520,12 +510,28 @@ public: * @param[in] im_width (Optional) Boxes whose centers (on the x axis) is beyond im_width will be filtered. Defaults to 1 * @param[in] im_height (Optional) Boxes whose centers (on the y axis) is beyond im_height will be filtered. Defaults to 1 */ - BoxNMSLimitInfo(float score_thresh = 0.05f, float nms = 0.3f, - int detections = 100, bool soft_nms_enabled = false, - NMSType soft_nms_method = NMSType::LINEAR, - float soft_nms_sigma = 0.5f, float soft_nms_min_score_thres = 0.001f, bool suppress_size = false, float min_size = 1.0f, float im_width = 1.0f, float im_height = 1.0f) - : _score_thresh(score_thresh), _nms(nms), _detections_per_im(detections), _soft_nms_enabled(soft_nms_enabled), _soft_nms_method(soft_nms_method), _soft_nms_sigma(soft_nms_sigma), - _soft_nms_min_score_thres(soft_nms_min_score_thres), _suppress_size(suppress_size), _min_size(min_size), _im_width(im_width), _im_height(im_height) + BoxNMSLimitInfo(float score_thresh = 0.05f, + float nms = 0.3f, + int detections = 100, + bool soft_nms_enabled = false, + NMSType soft_nms_method = NMSType::LINEAR, + float soft_nms_sigma = 0.5f, + float soft_nms_min_score_thres = 0.001f, + bool suppress_size = false, + float min_size = 1.0f, + float im_width = 1.0f, + float im_height = 1.0f) + : _score_thresh(score_thresh), + _nms(nms), + _detections_per_im(detections), + _soft_nms_enabled(soft_nms_enabled), + _soft_nms_method(soft_nms_method), + _soft_nms_sigma(soft_nms_sigma), + _soft_nms_min_score_thres(soft_nms_min_score_thres), + _suppress_size(suppress_size), + _min_size(min_size), + _im_width(im_width), + _im_height(im_height) { } /** Get the score threshold */ @@ -603,14 +609,13 @@ private: struct Padding2D { Padding2D() = default; - Padding2D(size_t left, size_t right, size_t top, size_t bottom) - : left(left), right(right), top(top), bottom(bottom) + Padding2D(size_t left, size_t right, size_t top, size_t bottom) : left(left), right(right), top(top), bottom(bottom) { } - size_t left = { 0 }; /**< Padding across the width dimension on the left, in elements. */ - size_t right = { 0 }; /**< Padding across the width dimension on the right, in elements. */ - size_t top = { 0 }; /**< Padding across the height dimension on the top, in elements. */ - size_t bottom = { 0 }; /**< Padding across the height dimension on the bottom, in elements. */ + size_t left = {0}; /**< Padding across the width dimension on the left, in elements. */ + size_t right = {0}; /**< Padding across the width dimension on the right, in elements. */ + size_t top = {0}; /**< Padding across the height dimension on the top, in elements. */ + size_t bottom = {0}; /**< Padding across the height dimension on the bottom, in elements. */ }; /** Padding information for 3D operations like Conv3d */ @@ -630,12 +635,12 @@ struct Padding3D { } - size_t left = { 0 }; /**< Padding across the width dimenstion on the left, in elements. */ - size_t right = { 0 }; /**< Padding across the width dimenstion on the right, in elements. */ - size_t top = { 0 }; /**< Padding across the height dimenstion on the top, in elements. */ - size_t bottom = { 0 }; /**< Padding across the height dimenstion on the bottom, in elements. */ - size_t front = { 0 }; /**< Padding across the depth dimenstion on the front, in elements. */ - size_t back = { 0 }; /**< Padding across the depth dimenstion on the back, in elements. */ + size_t left = {0}; /**< Padding across the width dimenstion on the left, in elements. */ + size_t right = {0}; /**< Padding across the width dimenstion on the right, in elements. */ + size_t top = {0}; /**< Padding across the height dimenstion on the top, in elements. */ + size_t bottom = {0}; /**< Padding across the height dimenstion on the bottom, in elements. */ + size_t front = {0}; /**< Padding across the depth dimenstion on the front, in elements. */ + size_t back = {0}; /**< Padding across the depth dimenstion on the back, in elements. */ }; /** PriorBox layer info */ @@ -667,9 +672,15 @@ public: * @param[in] img_size (Optional) Image size. * @param[in] steps (Optional) Step values. */ - PriorBoxLayerInfo(const std::vector<float> &min_sizes, const std::vector<float> &variances, float offset, bool flip = true, bool clip = false, - const std::vector<float> &max_sizes = {}, const std::vector<float> &aspect_ratios = {}, - const Coordinates2D &img_size = Coordinates2D{ 0, 0 }, const std::array<float, 2> &steps = { { 0.f, 0.f } }) + PriorBoxLayerInfo(const std::vector<float> &min_sizes, + const std::vector<float> &variances, + float offset, + bool flip = true, + bool clip = false, + const std::vector<float> &max_sizes = {}, + const std::vector<float> &aspect_ratios = {}, + const Coordinates2D &img_size = Coordinates2D{0, 0}, + const std::array<float, 2> &steps = {{0.f, 0.f}}) : _min_sizes(min_sizes), _variances(variances), _offset(offset), @@ -681,22 +692,22 @@ public: _steps(steps) { _aspect_ratios.push_back(1.); - for(unsigned int i = 0; i < aspect_ratios.size(); ++i) + for (unsigned int i = 0; i < aspect_ratios.size(); ++i) { float ar = aspect_ratios[i]; bool already_exist = false; - for(auto ar_new : _aspect_ratios) + for (auto ar_new : _aspect_ratios) { - if(fabs(ar - ar_new) < 1e-6) + if (fabs(ar - ar_new) < 1e-6) { already_exist = true; break; } } - if(!already_exist) + if (!already_exist) { _aspect_ratios.push_back(ar); - if(flip) + if (flip) { _aspect_ratios.push_back(1.f / ar); } @@ -808,8 +819,16 @@ public: * @param[in] variance_encoded_in_target (Optional) If true, variance is encoded in target. Otherwise we need to adjust the predicted offset accordingly.Default set to false. * @param[in] eta (Optional) Eta. */ - DetectionOutputLayerInfo(int num_classes, bool share_location, DetectionOutputLayerCodeType code_type, int keep_top_k, float nms_threshold, int top_k = -1, int background_label_id = -1, - float confidence_threshold = std::numeric_limits<float>::lowest(), bool variance_encoded_in_target = false, float eta = 1) + DetectionOutputLayerInfo(int num_classes, + bool share_location, + DetectionOutputLayerCodeType code_type, + int keep_top_k, + float nms_threshold, + int top_k = -1, + int background_label_id = -1, + float confidence_threshold = std::numeric_limits<float>::lowest(), + bool variance_encoded_in_target = false, + float eta = 1) : _num_classes(num_classes), _share_location(share_location), _code_type(code_type), @@ -923,8 +942,15 @@ public: * @param[in] detection_per_class (Optional) Number of detection per class. Used in the Regular Non-Max-Suppression. Defaults to 100. * @param[in] dequantize_scores (Optional) If the scores need to be dequantized. Defaults to true. */ - DetectionPostProcessLayerInfo(unsigned int max_detections, unsigned int max_classes_per_detection, float nms_score_threshold, float iou_threshold, unsigned int num_classes, - std::array<float, 4> scales_values, bool use_regular_nms = false, unsigned int detection_per_class = 100, bool dequantize_scores = true) + DetectionPostProcessLayerInfo(unsigned int max_detections, + unsigned int max_classes_per_detection, + float nms_score_threshold, + float iou_threshold, + unsigned int num_classes, + std::array<float, 4> scales_values, + bool use_regular_nms = false, + unsigned int detection_per_class = 100, + bool dequantize_scores = true) : _max_detections(max_detections), _max_classes_per_detection(max_classes_per_detection), _nms_score_threshold(nms_score_threshold), @@ -1240,8 +1266,14 @@ public: * @param[in] spatial_scale Spatial scale to be applied to the ROI coordinates and dimensions. * @param[in] sampling_ratio Number of samples to include in each pooling region (if set to zero, a ceil(roi_dims/pooling_dims)) */ - ROIPoolingLayerInfo(unsigned int pooled_width, unsigned int pooled_height, float spatial_scale, unsigned int sampling_ratio = 0) - : _pooled_width(pooled_width), _pooled_height(pooled_height), _spatial_scale(spatial_scale), _sampling_ratio(sampling_ratio) + ROIPoolingLayerInfo(unsigned int pooled_width, + unsigned int pooled_height, + float spatial_scale, + unsigned int sampling_ratio = 0) + : _pooled_width(pooled_width), + _pooled_height(pooled_height), + _spatial_scale(spatial_scale), + _sampling_ratio(sampling_ratio) { } /** Get the pooled width of the layer */ @@ -1288,10 +1320,24 @@ public: * @param[in] min_size (Optional)Size used to validate the anchors produced. Defaults to 16. * @param[in] values_per_roi (Optional)Values used to represent a ROI(Region of interest). Defaults to 4. */ - GenerateProposalsInfo(float im_width, float im_height, float im_scale, float spatial_scale = 1.0, int pre_nms_topN = 6000, int post_nms_topN = 300, float nms_thres = 0.7, float min_size = 16.0, + GenerateProposalsInfo(float im_width, + float im_height, + float im_scale, + float spatial_scale = 1.0, + int pre_nms_topN = 6000, + int post_nms_topN = 300, + float nms_thres = 0.7, + float min_size = 16.0, size_t values_per_roi = 4) - : _im_height(im_height), _im_width(im_width), _im_scale(im_scale), _spatial_scale(spatial_scale), _pre_nms_topN(pre_nms_topN), _post_nms_topN(post_nms_topN), _nms_thres(nms_thres), - _min_size(min_size), _values_per_roi(values_per_roi) + : _im_height(im_height), + _im_width(im_width), + _im_scale(im_scale), + _spatial_scale(spatial_scale), + _pre_nms_topN(pre_nms_topN), + _post_nms_topN(post_nms_topN), + _nms_thres(nms_thres), + _min_size(min_size), + _values_per_roi(values_per_roi) { } @@ -1417,11 +1463,20 @@ public: * @param[in] correct_transform_coords (Optional)Correct bounding box transform coordinates. Defaults to false * @param[in] bbox_xform_clip (Optional)Minimum bounding box width and height after bounding box transformation in log-space. Defaults to log(1000/16) */ - BoundingBoxTransformInfo(float img_width, float img_height, float scale, bool apply_scale = false, const std::array<float, 4> weights = { { 1.f, 1.f, 1.f, 1.f } }, bool correct_transform_coords = - false, - float bbox_xform_clip = - 4.135166556742356f) - : _img_width(img_width), _img_height(img_height), _scale(scale), _apply_scale(apply_scale), _correct_transform_coords(correct_transform_coords), _weights(weights), _bbox_xform_clip(bbox_xform_clip) + BoundingBoxTransformInfo(float img_width, + float img_height, + float scale, + bool apply_scale = false, + const std::array<float, 4> weights = {{1.f, 1.f, 1.f, 1.f}}, + bool correct_transform_coords = false, + float bbox_xform_clip = 4.135166556742356f) + : _img_width(img_width), + _img_height(img_height), + _scale(scale), + _apply_scale(apply_scale), + _correct_transform_coords(correct_transform_coords), + _weights(weights), + _bbox_xform_clip(bbox_xform_clip) { } @@ -1484,7 +1539,12 @@ public: * @param[in] is_scaled (Optional) Boolean that specifies if alpha will be scaled by the normalization size or not. * Should be false to follow [Krichevksy 2012]. */ - NormalizationLayerInfo(NormType type, uint32_t norm_size = 5, float alpha = 0.0001f, float beta = 0.5f, float kappa = 1.f, bool is_scaled = true) + NormalizationLayerInfo(NormType type, + uint32_t norm_size = 5, + float alpha = 0.0001f, + float beta = 0.5f, + float kappa = 1.f, + bool is_scaled = true) : _type(type), _norm_size(norm_size), _alpha(alpha), _beta(beta), _kappa(kappa), _is_scaled(is_scaled) { } @@ -1612,7 +1672,12 @@ class WeightsInfo public: /** Default constructor */ WeightsInfo() - : _are_reshaped(false), _kernel_width(0), _kernel_height(0), _num_kernels(0), _retain_internal_weights(false), _weight_format(arm_compute::WeightFormat::UNSPECIFIED) + : _are_reshaped(false), + _kernel_width(0), + _kernel_height(0), + _num_kernels(0), + _retain_internal_weights(false), + _weight_format(arm_compute::WeightFormat::UNSPECIFIED) { } /** Constructor @@ -1624,9 +1689,18 @@ public: * @param[in] retain_internal_weights (Optional) True if internal reshaped weights must be retained. Used for reconfiguration purposes. Default is false. * @param[in] weight_format (Optional) arm_gemm:WeightFormat enumeration requested by the user. Default is arm_compute::WeightFormat::UNSPECIFIED. */ - WeightsInfo(bool are_reshaped, unsigned int kernel_width, unsigned int kernel_height, unsigned int num_kernels, bool retain_internal_weights = false, - arm_compute::WeightFormat weight_format = arm_compute::WeightFormat::UNSPECIFIED) - : _are_reshaped(are_reshaped), _kernel_width(kernel_width), _kernel_height(kernel_height), _num_kernels(num_kernels), _retain_internal_weights(retain_internal_weights), _weight_format(weight_format) + WeightsInfo(bool are_reshaped, + unsigned int kernel_width, + unsigned int kernel_height, + unsigned int num_kernels, + bool retain_internal_weights = false, + arm_compute::WeightFormat weight_format = arm_compute::WeightFormat::UNSPECIFIED) + : _are_reshaped(are_reshaped), + _kernel_width(kernel_width), + _kernel_height(kernel_height), + _num_kernels(num_kernels), + _retain_internal_weights(retain_internal_weights), + _weight_format(weight_format) { } /** Flag which specifies if the weights tensor has been reshaped. @@ -1698,7 +1772,14 @@ class GEMMReshapeInfo final public: /** Default constructor */ GEMMReshapeInfo() - : _m(1), _n(1), _k(1), _mult_transpose1xW_width(1), _mult_interleave4x4_height(1), _depth_output_gemm3d(0), _reinterpret_input_as_3d(false), _broadcast_bias(false) + : _m(1), + _n(1), + _k(1), + _mult_transpose1xW_width(1), + _mult_interleave4x4_height(1), + _depth_output_gemm3d(0), + _reinterpret_input_as_3d(false), + _broadcast_bias(false) { } /** Constructor @@ -1714,9 +1795,22 @@ public: * to perform 1x1 convolutions with the NHWC data layout) * @param[in] broadcast_bias (Optional) Broadcast the shape of the bias tensor from a vector to a matrix. */ - GEMMReshapeInfo(int m, int n, int k, int mult_transpose1xW_width = 1, int mult_interleave4x4_height = 1, int depth_output_gemm3d = 0, bool reinterpret_input_as_3d = false, bool broadcast_bias = false) - : _m(m), _n(n), _k(k), _mult_transpose1xW_width(mult_transpose1xW_width), _mult_interleave4x4_height(mult_interleave4x4_height), _depth_output_gemm3d(depth_output_gemm3d), - _reinterpret_input_as_3d(reinterpret_input_as_3d), _broadcast_bias(broadcast_bias) + GEMMReshapeInfo(int m, + int n, + int k, + int mult_transpose1xW_width = 1, + int mult_interleave4x4_height = 1, + int depth_output_gemm3d = 0, + bool reinterpret_input_as_3d = false, + bool broadcast_bias = false) + : _m(m), + _n(n), + _k(k), + _mult_transpose1xW_width(mult_transpose1xW_width), + _mult_interleave4x4_height(mult_interleave4x4_height), + _depth_output_gemm3d(depth_output_gemm3d), + _reinterpret_input_as_3d(reinterpret_input_as_3d), + _broadcast_bias(broadcast_bias) { } /** Number of matrix A rows @@ -1806,11 +1900,11 @@ struct GEMMLHSMatrixInfo : m0(m), k0(k), v0(v), transpose(trans), interleave(inter) { } - unsigned int m0{ 1 }; /**< Number of rows processed by the matrix multiplication */ - unsigned int k0{ 1 }; /**< Number of partial accumulations performed by the matrix multiplication */ - unsigned int v0{ 1 }; /**< Number of vertical blocks of size (m0xk0) stored on the same output row */ - bool transpose{ true }; /**< True if the (m0xk0) block has to be transposed before been stored */ - bool interleave{ true }; /**< True if the v0 (m0xk0) blocks have to be interleaved in the output row */ + unsigned int m0{1}; /**< Number of rows processed by the matrix multiplication */ + unsigned int k0{1}; /**< Number of partial accumulations performed by the matrix multiplication */ + unsigned int v0{1}; /**< Number of vertical blocks of size (m0xk0) stored on the same output row */ + bool transpose{true}; /**< True if the (m0xk0) block has to be transposed before been stored */ + bool interleave{true}; /**< True if the v0 (m0xk0) blocks have to be interleaved in the output row */ }; /** GEMM RHS (Right Hand Side) matrix information */ @@ -1821,12 +1915,13 @@ struct GEMMRHSMatrixInfo : n0(n), k0(k), h0(h), transpose(trans), interleave(inter), export_to_cl_image(export_to_cl_img) { } - unsigned int n0{ 1 }; /**< Number of columns processed by the matrix multiplication */ - unsigned int k0{ 1 }; /**< Number of partial accumulations performed by the matrix multiplication */ - unsigned int h0{ 1 }; /**< Number of horizontal blocks of size (k0xn0) stored on the same output row */ - bool transpose{ true }; /**< True if the (k0xn0) block has to be transposed before been stored */ - bool interleave{ true }; /**< True if the h0 (k0xn0) blocks have to be interleaved in the output row */ - bool export_to_cl_image{ false }; /**< True if the reshaped rhs has to be exported to cl_image. n0 must be equal to 4 */ + unsigned int n0{1}; /**< Number of columns processed by the matrix multiplication */ + unsigned int k0{1}; /**< Number of partial accumulations performed by the matrix multiplication */ + unsigned int h0{1}; /**< Number of horizontal blocks of size (k0xn0) stored on the same output row */ + bool transpose{true}; /**< True if the (k0xn0) block has to be transposed before been stored */ + bool interleave{true}; /**< True if the h0 (k0xn0) blocks have to be interleaved in the output row */ + bool export_to_cl_image{ + false}; /**< True if the reshaped rhs has to be exported to cl_image. n0 must be equal to 4 */ }; class ITensorInfo; @@ -1842,16 +1937,23 @@ struct WinogradInfo * @param[in] conv_info Convolution info (Pads, strides) * @param[in] data_layout Data layout to use for the output tensor once the convolution has been applied */ - WinogradInfo(Size2D output_tile_sz, Size2D kernel_sz, Size2D input_dims, PadStrideInfo conv_info, DataLayout data_layout) - : output_tile_size(output_tile_sz), kernel_size(kernel_sz), input_dimensions(input_dims), convolution_info(conv_info), output_data_layout(data_layout) - { - } - - Size2D output_tile_size{}; /**< Width and height of the output tile */ - Size2D kernel_size{}; /**< Width and height of the kernel*/ - Size2D input_dimensions{}; /**< Width and height of the input tensor before the convolution is applied */ - PadStrideInfo convolution_info{}; /**< Convolution info (Pads, strides,...) */ - DataLayout output_data_layout{ DataLayout::NCHW }; /**< Data layout to use for the output tensor once the convolution has been applied (NCHW or NHWC) */ + WinogradInfo( + Size2D output_tile_sz, Size2D kernel_sz, Size2D input_dims, PadStrideInfo conv_info, DataLayout data_layout) + : output_tile_size(output_tile_sz), + kernel_size(kernel_sz), + input_dimensions(input_dims), + convolution_info(conv_info), + output_data_layout(data_layout) + { + } + + Size2D output_tile_size{}; /**< Width and height of the output tile */ + Size2D kernel_size{}; /**< Width and height of the kernel*/ + Size2D input_dimensions{}; /**< Width and height of the input tensor before the convolution is applied */ + PadStrideInfo convolution_info{}; /**< Convolution info (Pads, strides,...) */ + DataLayout output_data_layout{ + DataLayout:: + NCHW}; /**< Data layout to use for the output tensor once the convolution has been applied (NCHW or NHWC) */ }; /** IO formatting information class*/ diff --git a/arm_compute/core/Utils.h b/arm_compute/core/Utils.h index c5b50167bf..a2146522f7 100644 --- a/arm_compute/core/Utils.h +++ b/arm_compute/core/Utils.h @@ -69,7 +69,7 @@ template <typename T> inline void permute_strides(Dimensions<T> &dimensions, const PermutationVector &perm) { const auto old_dim = utility::make_array<Dimensions<T>::num_max_dimensions>(dimensions.begin(), dimensions.end()); - for(unsigned int i = 0; i < perm.num_dimensions(); ++i) + for (unsigned int i = 0; i < perm.num_dimensions(); ++i) { T dimension_val = old_dim[i]; dimensions.set(perm[i], dimension_val); @@ -87,7 +87,11 @@ inline void permute_strides(Dimensions<T> &dimensions, const PermutationVector & * * @return PadStrideInfo for SAME padding */ -PadStrideInfo calculate_same_pad(TensorShape input_shape, TensorShape weights_shape, PadStrideInfo conv_info, DataLayout data_layout = DataLayout::NCHW, const Size2D &dilation = Size2D(1u, 1u), +PadStrideInfo calculate_same_pad(TensorShape input_shape, + TensorShape weights_shape, + PadStrideInfo conv_info, + DataLayout data_layout = DataLayout::NCHW, + const Size2D &dilation = Size2D(1u, 1u), const DimensionRoundingType &rounding_type = DimensionRoundingType::FLOOR); /** Returns expected width and height of the deconvolution's output tensor. @@ -100,8 +104,10 @@ PadStrideInfo calculate_same_pad(TensorShape input_shape, TensorShape weights_sh * * @return A pair with the new width in the first position and the new height in the second. */ -std::pair<unsigned int, unsigned int> deconvolution_output_dimensions(unsigned int in_width, unsigned int in_height, - unsigned int kernel_width, unsigned int kernel_height, +std::pair<unsigned int, unsigned int> deconvolution_output_dimensions(unsigned int in_width, + unsigned int in_height, + unsigned int kernel_width, + unsigned int kernel_height, const PadStrideInfo &pad_stride_info); /** Returns expected width and height of output scaled tensor depending on dimensions rounding mode. @@ -115,8 +121,10 @@ std::pair<unsigned int, unsigned int> deconvolution_output_dimensions(unsigned i * * @return A pair with the new width in the first position and the new height in the second. */ -std::pair<unsigned int, unsigned int> scaled_dimensions(int width, int height, - int kernel_width, int kernel_height, +std::pair<unsigned int, unsigned int> scaled_dimensions(int width, + int height, + int kernel_width, + int kernel_height, const PadStrideInfo &pad_stride_info, const Size2D &dilation = Size2D(1U, 1U)); @@ -130,9 +138,8 @@ std::pair<unsigned int, unsigned int> scaled_dimensions(int width, int height, * * @return A pair with the new width in the first position and the new height in the second, returned values can be < 1 */ -std::pair<int, int> scaled_dimensions_signed(int width, int height, - int kernel_width, int kernel_height, - const PadStrideInfo &pad_stride_info); +std::pair<int, int> scaled_dimensions_signed( + int width, int height, int kernel_width, int kernel_height, const PadStrideInfo &pad_stride_info); /** Returns calculated width, height and depth of output scaled tensor depending on dimensions rounding mode. * @@ -147,8 +154,12 @@ std::pair<int, int> scaled_dimensions_signed(int width, int height, * @return A tuple with the new width in the first position, the new height in the second, and the new depth in the third. * Returned values can be < 1 */ -std::tuple<int, int, int> scaled_3d_dimensions_signed(int width, int height, int depth, - int kernel_width, int kernel_height, int kernel_depth, +std::tuple<int, int, int> scaled_3d_dimensions_signed(int width, + int height, + int depth, + int kernel_width, + int kernel_height, + int kernel_depth, const Pooling3dLayerInfo &pool3d_info); /** Check if the given reduction operation should be handled in a serial way. @@ -178,7 +189,9 @@ QuantizationInfo get_softmax_output_quantization_info(DataType input_type, bool * * @return The pair with minimum and maximum values */ -std::pair<int32_t, int32_t> get_quantized_activation_min_max(const ActivationLayerInfo &act_info, DataType data_type, UniformQuantizationInfo oq_info); +std::pair<int32_t, int32_t> get_quantized_activation_min_max(const ActivationLayerInfo &act_info, + DataType data_type, + UniformQuantizationInfo oq_info); /** Convert a channel identity into a string. * @@ -295,26 +308,27 @@ inline size_t num_of_elements_in_range(const float start, const float end, const * @param[in] element_delim (Optional) Delimeter among the consecutive elements. Defaults to space delimeter */ template <typename T> -void print_consecutive_elements_impl(std::ostream &s, const T *ptr, unsigned int n, int stream_width = 0, const std::string &element_delim = " ") +void print_consecutive_elements_impl( + std::ostream &s, const T *ptr, unsigned int n, int stream_width = 0, const std::string &element_delim = " ") { using print_type = typename std::conditional<std::is_floating_point<T>::value, T, int>::type; std::ios stream_status(nullptr); stream_status.copyfmt(s); - for(unsigned int i = 0; i < n; ++i) + for (unsigned int i = 0; i < n; ++i) { // Set stream width as it is not a "sticky" stream manipulator - if(stream_width != 0) + if (stream_width != 0) { s.width(stream_width); } - if(std::is_same<typename std::decay<T>::type, half>::value) + if (std::is_same<typename std::decay<T>::type, half>::value) { // We use T instead of print_type here is because the std::is_floating_point<half> returns false and then the print_type becomes int. s << std::right << static_cast<T>(ptr[i]) << element_delim; } - else if(std::is_same<typename std::decay<T>::type, bfloat16>::value) + else if (std::is_same<typename std::decay<T>::type, bfloat16>::value) { // We use T instead of print_type here is because the std::is_floating_point<bfloat16> returns false and then the print_type becomes int. s << std::right << float(ptr[i]) << element_delim; @@ -343,17 +357,17 @@ int max_consecutive_elements_display_width_impl(std::ostream &s, const T *ptr, u using print_type = typename std::conditional<std::is_floating_point<T>::value, T, int>::type; int max_width = -1; - for(unsigned int i = 0; i < n; ++i) + for (unsigned int i = 0; i < n; ++i) { std::stringstream ss; ss.copyfmt(s); - if(std::is_same<typename std::decay<T>::type, half>::value) + if (std::is_same<typename std::decay<T>::type, half>::value) { // We use T instead of print_type here is because the std::is_floating_point<half> returns false and then the print_type becomes int. ss << static_cast<T>(ptr[i]); } - else if(std::is_same<typename std::decay<T>::type, bfloat16>::value) + else if (std::is_same<typename std::decay<T>::type, bfloat16>::value) { // We use T instead of print_type here is because the std::is_floating_point<bfloat> returns false and then the print_type becomes int. ss << float(ptr[i]); @@ -377,7 +391,12 @@ int max_consecutive_elements_display_width_impl(std::ostream &s, const T *ptr, u * @param[in] stream_width (Optional) Width of the stream. If set to 0 the element's width is used. Defaults to 0. * @param[in] element_delim (Optional) Delimeter among the consecutive elements. Defaults to space delimeter */ -void print_consecutive_elements(std::ostream &s, DataType dt, const uint8_t *ptr, unsigned int n, int stream_width, const std::string &element_delim = " "); +void print_consecutive_elements(std::ostream &s, + DataType dt, + const uint8_t *ptr, + unsigned int n, + int stream_width, + const std::string &element_delim = " "); /** Identify the maximum width of n consecutive elements. * @@ -390,5 +409,5 @@ void print_consecutive_elements(std::ostream &s, DataType dt, const uint8_t *ptr */ int max_consecutive_elements_display_width(std::ostream &s, DataType dt, const uint8_t *ptr, unsigned int n); #endif /* ARM_COMPUTE_ASSERTS_ENABLED */ -} +} // namespace arm_compute #endif /*ARM_COMPUTE_UTILS_H */ diff --git a/arm_compute/core/Validate.h b/arm_compute/core/Validate.h index 5bffc16f3b..5550560aff 100644 --- a/arm_compute/core/Validate.h +++ b/arm_compute/core/Validate.h @@ -24,13 +24,13 @@ #ifndef ARM_COMPUTE_VALIDATE_H #define ARM_COMPUTE_VALIDATE_H -#include "arm_compute/core/utils/DataLayoutUtils.h" -#include "arm_compute/core/utils/DataTypeUtils.h" #include "arm_compute/core/Error.h" -#include "arm_compute/core/utils/FormatUtils.h" #include "arm_compute/core/IKernel.h" #include "arm_compute/core/ITensor.h" #include "arm_compute/core/QuantizationInfo.h" +#include "arm_compute/core/utils/DataLayoutUtils.h" +#include "arm_compute/core/utils/DataTypeUtils.h" +#include "arm_compute/core/utils/FormatUtils.h" #include "arm_compute/core/Window.h" #include <algorithm> @@ -50,9 +50,9 @@ namespace detail template <typename T> inline bool have_different_dimensions(const Dimensions<T> &dim1, const Dimensions<T> &dim2, unsigned int upper_dim) { - for(unsigned int i = upper_dim; i < arm_compute::Dimensions<T>::num_max_dimensions; ++i) + for (unsigned int i = upper_dim; i < arm_compute::Dimensions<T>::num_max_dimensions; ++i) { - if(dim1[i] != dim2[i]) + if (dim1[i] != dim2[i]) { return true; } @@ -80,7 +80,7 @@ public: * @param[in] line Source code line. Used for error reporting. */ compare_dimension(const Dimensions<T> &dim, const char *function, const char *file, int line) - : _dim{ dim }, _function{ function }, _file{ file }, _line{ line } + : _dim{dim}, _function{function}, _file{file}, _line{line} { } @@ -111,7 +111,7 @@ inline arm_compute::Status for_each_error(F &&) } template <typename F, typename T, typename... Ts> -inline arm_compute::Status for_each_error(F &&func, T &&arg, Ts &&... args) +inline arm_compute::Status for_each_error(F &&func, T &&arg, Ts &&...args) { ARM_COMPUTE_RETURN_ON_ERROR(func(arg)); ARM_COMPUTE_RETURN_ON_ERROR(for_each_error(func, args...)); @@ -148,13 +148,11 @@ struct get_tensor_info_t<ITensorInfo *> * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_nullptr(const char *function, const char *file, const int line, Ts &&... pointers) +inline arm_compute::Status error_on_nullptr(const char *function, const char *file, const int line, Ts &&...pointers) { - const std::array<const void *, sizeof...(Ts)> pointers_array{ { std::forward<Ts>(pointers)... } }; - bool has_nullptr = std::any_of(pointers_array.begin(), pointers_array.end(), [&](const void *ptr) - { - return (ptr == nullptr); - }); + const std::array<const void *, sizeof...(Ts)> pointers_array{{std::forward<Ts>(pointers)...}}; + bool has_nullptr = + std::any_of(pointers_array.begin(), pointers_array.end(), [&](const void *ptr) { return (ptr == nullptr); }); ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(has_nullptr, function, file, line, "Nullptr object!"); return arm_compute::Status{}; } @@ -178,8 +176,8 @@ inline arm_compute::Status error_on_nullptr(const char *function, const char *fi * * @return Status */ -arm_compute::Status error_on_mismatching_windows(const char *function, const char *file, const int line, - const Window &full, const Window &win); +arm_compute::Status error_on_mismatching_windows( + const char *function, const char *file, const int line, const Window &full, const Window &win); #define ARM_COMPUTE_ERROR_ON_MISMATCHING_WINDOWS(f, w) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_mismatching_windows(__func__, __FILE__, __LINE__, f, w)) #define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_WINDOWS(f, w) \ @@ -200,8 +198,8 @@ arm_compute::Status error_on_mismatching_windows(const char *function, const cha * * @return Status */ -arm_compute::Status error_on_invalid_subwindow(const char *function, const char *file, const int line, - const Window &full, const Window &sub); +arm_compute::Status error_on_invalid_subwindow( + const char *function, const char *file, const int line, const Window &full, const Window &sub); #define ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(f, s) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_invalid_subwindow(__func__, __FILE__, __LINE__, f, s)) #define ARM_COMPUTE_RETURN_ERROR_ON_INVALID_SUBWINDOW(f, s) \ @@ -220,12 +218,14 @@ arm_compute::Status error_on_invalid_subwindow(const char *function, const char * * @return Status */ -arm_compute::Status error_on_window_not_collapsable_at_dimension(const char *function, const char *file, const int line, - const Window &full, const Window &window, const int dim); +arm_compute::Status error_on_window_not_collapsable_at_dimension( + const char *function, const char *file, const int line, const Window &full, const Window &window, const int dim); #define ARM_COMPUTE_ERROR_ON_WINDOW_NOT_COLLAPSABLE_AT_DIMENSION(f, w, d) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_window_not_collapsable_at_dimension(__func__, __FILE__, __LINE__, f, w, d)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_window_not_collapsable_at_dimension(__func__, __FILE__, __LINE__, f, w, d)) #define ARM_COMPUTE_RETURN_ERROR_ON_WINDOW_NOT_COLLAPSABLE_AT_DIMENSION(f, w, d) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_window_not_collapsable_at_dimension(__func__, __FILE__, __LINE__, f, w, d)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_window_not_collapsable_at_dimension(__func__, __FILE__, __LINE__, f, w, d)) /** Return an error if the passed coordinates have too many dimensions. * @@ -239,8 +239,8 @@ arm_compute::Status error_on_window_not_collapsable_at_dimension(const char *fun * * @return Status */ -arm_compute::Status error_on_coordinates_dimensions_gte(const char *function, const char *file, const int line, - const Coordinates &pos, unsigned int max_dim); +arm_compute::Status error_on_coordinates_dimensions_gte( + const char *function, const char *file, const int line, const Coordinates &pos, unsigned int max_dim); #define ARM_COMPUTE_ERROR_ON_COORDINATES_DIMENSIONS_GTE(p, md) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_coordinates_dimensions_gte(__func__, __FILE__, __LINE__, p, md)) #define ARM_COMPUTE_RETURN_ERROR_ON_COORDINATES_DIMENSIONS_GTE(p, md) \ @@ -258,8 +258,8 @@ arm_compute::Status error_on_coordinates_dimensions_gte(const char *function, co * * @return Status */ -arm_compute::Status error_on_window_dimensions_gte(const char *function, const char *file, const int line, - const Window &win, unsigned int max_dim); +arm_compute::Status error_on_window_dimensions_gte( + const char *function, const char *file, const int line, const Window &win, unsigned int max_dim); #define ARM_COMPUTE_ERROR_ON_WINDOW_DIMENSIONS_GTE(w, md) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_window_dimensions_gte(__func__, __FILE__, __LINE__, w, md)) #define ARM_COMPUTE_RETURN_ERROR_ON_WINDOW_DIMENSIONS_GTE(w, md) \ @@ -277,16 +277,23 @@ arm_compute::Status error_on_window_dimensions_gte(const char *function, const c * @return Status */ template <typename T, typename... Ts> -arm_compute::Status error_on_mismatching_dimensions(const char *function, const char *file, int line, - const Dimensions<T> &dim1, const Dimensions<T> &dim2, Ts &&... dims) +arm_compute::Status error_on_mismatching_dimensions(const char *function, + const char *file, + int line, + const Dimensions<T> &dim1, + const Dimensions<T> &dim2, + Ts &&...dims) { - ARM_COMPUTE_RETURN_ON_ERROR(detail::for_each_error(detail::compare_dimension<T>(dim1, function, file, line), dim2, std::forward<Ts>(dims)...)); + ARM_COMPUTE_RETURN_ON_ERROR(detail::for_each_error(detail::compare_dimension<T>(dim1, function, file, line), dim2, + std::forward<Ts>(dims)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_MISMATCHING_DIMENSIONS(...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_mismatching_dimensions(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_mismatching_dimensions(__func__, __FILE__, __LINE__, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DIMENSIONS(...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_dimensions(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_mismatching_dimensions(__func__, __FILE__, __LINE__, __VA_ARGS__)) /** Return true if the given format has horizontal subsampling. * @@ -296,7 +303,10 @@ arm_compute::Status error_on_mismatching_dimensions(const char *function, const */ inline bool has_format_horizontal_subsampling(Format format) { - return (format == Format::YUYV422 || format == Format::UYVY422 || format == Format::NV12 || format == Format::NV21 || format == Format::IYUV || format == Format::UV88) ? true : false; + return (format == Format::YUYV422 || format == Format::UYVY422 || format == Format::NV12 || + format == Format::NV21 || format == Format::IYUV || format == Format::UV88) + ? true + : false; } /** Return true if the given format has vertical subsampling. @@ -307,7 +317,9 @@ inline bool has_format_horizontal_subsampling(Format format) */ inline bool has_format_vertical_subsampling(Format format) { - return (format == Format::NV12 || format == Format::NV21 || format == Format::IYUV || format == Format::UV88) ? true : false; + return (format == Format::NV12 || format == Format::NV21 || format == Format::IYUV || format == Format::UV88) + ? true + : false; } /** Adjust tensor shape size if width or height are odd for a given multi-planar format. No modification is done for other formats. @@ -325,16 +337,16 @@ inline bool has_format_vertical_subsampling(Format format) */ inline TensorShape adjust_odd_shape(const TensorShape &shape, Format format) { - TensorShape output{ shape }; + TensorShape output{shape}; // Force width to be even for formats which require subsampling of the U and V channels - if(has_format_horizontal_subsampling(format)) + if (has_format_horizontal_subsampling(format)) { output.set(0, (output.x() + 1) & ~1U); } // Force height to be even for formats which require subsampling of the U and V channels - if(has_format_vertical_subsampling(format)) + if (has_format_vertical_subsampling(format)) { output.set(1, (output.y() + 1) & ~1U); } @@ -354,18 +366,20 @@ inline TensorShape adjust_odd_shape(const TensorShape &shape, Format format) * @return Status */ template <typename... Ts> -arm_compute::Status error_on_tensors_not_even(const char *function, const char *file, int line, - const Format &format, const ITensor *tensor1, Ts... tensors) +arm_compute::Status error_on_tensors_not_even( + const char *function, const char *file, int line, const Format &format, const ITensor *tensor1, Ts... tensors) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor1 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, std::forward<Ts>(tensors)...)); - const std::array < const ITensor *, 1 + sizeof...(Ts) > tensors_info_array{ { tensor1, std::forward<Ts>(tensors)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_info_array.cbegin(), tensors_info_array.cend(), [&](const ITensor * tensor) - { - const TensorShape correct_shape = adjust_odd_shape(tensor->info()->tensor_shape(), format); - return detail::have_different_dimensions(tensor->info()->tensor_shape(), correct_shape, 2); - }), - function, file, line, "Tensor shape has odd dimensions"); + const std::array<const ITensor *, 1 + sizeof...(Ts)> tensors_info_array{{tensor1, std::forward<Ts>(tensors)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG( + std::any_of(tensors_info_array.cbegin(), tensors_info_array.cend(), + [&](const ITensor *tensor) + { + const TensorShape correct_shape = adjust_odd_shape(tensor->info()->tensor_shape(), format); + return detail::have_different_dimensions(tensor->info()->tensor_shape(), correct_shape, 2); + }), + function, file, line, "Tensor shape has odd dimensions"); return arm_compute::Status{}; } @@ -382,21 +396,22 @@ arm_compute::Status error_on_tensors_not_even(const char *function, const char * * * @return The subsampled tensor shape. */ -inline TensorShape calculate_subsampled_shape(const TensorShape &shape, Format format, Channel channel = Channel::UNKNOWN) +inline TensorShape +calculate_subsampled_shape(const TensorShape &shape, Format format, Channel channel = Channel::UNKNOWN) { - TensorShape output{ shape }; + TensorShape output{shape}; // Subsample shape only for U or V channel - if(Channel::U == channel || Channel::V == channel || Channel::UNKNOWN == channel) + if (Channel::U == channel || Channel::V == channel || Channel::UNKNOWN == channel) { // Subsample width for the tensor shape when channel is U or V - if(has_format_horizontal_subsampling(format)) + if (has_format_horizontal_subsampling(format)) { output.set(0, output.x() / 2U); } // Subsample height for the tensor shape when channel is U or V - if(has_format_vertical_subsampling(format)) + if (has_format_vertical_subsampling(format)) { output.set(1, output.y() / 2U); } @@ -418,25 +433,32 @@ inline TensorShape calculate_subsampled_shape(const TensorShape &shape, Format f * @return Status */ template <typename... Ts> -arm_compute::Status error_on_tensors_not_subsampled(const char *function, const char *file, int line, - const Format &format, const TensorShape &shape, const ITensor *tensor1, Ts... tensors) +arm_compute::Status error_on_tensors_not_subsampled(const char *function, + const char *file, + int line, + const Format &format, + const TensorShape &shape, + const ITensor *tensor1, + Ts... tensors) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor1 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, std::forward<Ts>(tensors)...)); - const TensorShape sub2_shape = calculate_subsampled_shape(shape, format); - const std::array < const ITensor *, 1 + sizeof...(Ts) > tensors_info_array{ { tensor1, std::forward<Ts>(tensors)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_info_array.cbegin(), tensors_info_array.cend(), [&](const ITensor * tensor) - { - return detail::have_different_dimensions(tensor->info()->tensor_shape(), sub2_shape, 2); - }), - function, file, line, "Tensor shape has mismatch dimensions for sub-sampling"); + const TensorShape sub2_shape = calculate_subsampled_shape(shape, format); + const std::array<const ITensor *, 1 + sizeof...(Ts)> tensors_info_array{{tensor1, std::forward<Ts>(tensors)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG( + std::any_of(tensors_info_array.cbegin(), tensors_info_array.cend(), + [&](const ITensor *tensor) + { return detail::have_different_dimensions(tensor->info()->tensor_shape(), sub2_shape, 2); }), + function, file, line, "Tensor shape has mismatch dimensions for sub-sampling"); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_TENSORS_NOT_SUBSAMPLED(...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_tensors_not_subsampled(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_tensors_not_subsampled(__func__, __FILE__, __LINE__, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_TENSORS_NOT_SUBSAMPLED(...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_tensors_not_subsampled(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_tensors_not_subsampled(__func__, __FILE__, __LINE__, __VA_ARGS__)) /** Return an error if the passed two tensor infos have different shapes from the given dimension * @@ -450,10 +472,15 @@ arm_compute::Status error_on_tensors_not_subsampled(const char *function, const * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_shapes(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos) +inline arm_compute::Status error_on_mismatching_shapes(const char *function, + const char *file, + const int line, + const ITensorInfo *tensor_info_1, + const ITensorInfo *tensor_info_2, + Ts... tensor_infos) { - return error_on_mismatching_shapes(function, file, line, 0U, tensor_info_1, tensor_info_2, std::forward<Ts>(tensor_infos)...); + return error_on_mismatching_shapes(function, file, line, 0U, tensor_info_1, tensor_info_2, + std::forward<Ts>(tensor_infos)...); } /** Return an error if the passed two tensors have different shapes from the given dimension * @@ -467,8 +494,12 @@ inline arm_compute::Status error_on_mismatching_shapes(const char *function, con * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_shapes(const char *function, const char *file, const int line, - const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors) +inline arm_compute::Status error_on_mismatching_shapes(const char *function, + const char *file, + const int line, + const ITensor *tensor_1, + const ITensor *tensor_2, + Ts... tensors) { return error_on_mismatching_shapes(function, file, line, 0U, tensor_1, tensor_2, std::forward<Ts>(tensors)...); } @@ -485,19 +516,28 @@ inline arm_compute::Status error_on_mismatching_shapes(const char *function, con * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_shapes(const char *function, const char *file, const int line, - unsigned int upper_dim, const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos) +inline arm_compute::Status error_on_mismatching_shapes(const char *function, + const char *file, + const int line, + unsigned int upper_dim, + const ITensorInfo *tensor_info_1, + const ITensorInfo *tensor_info_2, + Ts... tensor_infos) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info_1 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info_2 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, tensor_infos...)); - const std::array < const ITensorInfo *, 2 + sizeof...(Ts) > tensors_info_array{ { tensor_info_1, tensor_info_2, tensor_infos... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(std::next(tensors_info_array.cbegin()), tensors_info_array.cend(), [&](const ITensorInfo * tensor_info) - { - return detail::have_different_dimensions((*tensors_info_array.cbegin())->tensor_shape(), tensor_info->tensor_shape(), upper_dim); - }), - function, file, line, "Tensors have different shapes"); + const std::array<const ITensorInfo *, 2 + sizeof...(Ts)> tensors_info_array{ + {tensor_info_1, tensor_info_2, tensor_infos...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(std::next(tensors_info_array.cbegin()), tensors_info_array.cend(), + [&](const ITensorInfo *tensor_info) + { + return detail::have_different_dimensions( + (*tensors_info_array.cbegin())->tensor_shape(), + tensor_info->tensor_shape(), upper_dim); + }), + function, file, line, "Tensors have different shapes"); return arm_compute::Status{}; } /** Return an error if the passed two tensors have different shapes from the given dimension @@ -513,14 +553,20 @@ inline arm_compute::Status error_on_mismatching_shapes(const char *function, con * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_shapes(const char *function, const char *file, const int line, - unsigned int upper_dim, const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors) +inline arm_compute::Status error_on_mismatching_shapes(const char *function, + const char *file, + const int line, + unsigned int upper_dim, + const ITensor *tensor_1, + const ITensor *tensor_2, + Ts... tensors) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_1 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_2 == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, tensors...)); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_shapes(function, file, line, upper_dim, tensor_1->info(), tensor_2->info(), - detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); + ARM_COMPUTE_RETURN_ON_ERROR( + ::arm_compute::error_on_mismatching_shapes(function, file, line, upper_dim, tensor_1->info(), tensor_2->info(), + detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_MISMATCHING_SHAPES(...) \ @@ -539,19 +585,18 @@ inline arm_compute::Status error_on_mismatching_shapes(const char *function, con * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_data_layouts(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, Ts... tensor_infos) +inline arm_compute::Status error_on_mismatching_data_layouts( + const char *function, const char *file, const int line, const ITensorInfo *tensor_info, Ts... tensor_infos) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, tensor_infos...)); - DataLayout &&tensor_data_layout = tensor_info->data_layout(); - const std::array<const ITensorInfo *, sizeof...(Ts)> tensors_infos_array{ { tensor_infos... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_infos_array.begin(), tensors_infos_array.end(), [&](const ITensorInfo * tensor_info_obj) - { - return tensor_info_obj->data_layout() != tensor_data_layout; - }), - function, file, line, "Tensors have different data layouts"); + DataLayout &&tensor_data_layout = tensor_info->data_layout(); + const std::array<const ITensorInfo *, sizeof...(Ts)> tensors_infos_array{{tensor_infos...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_infos_array.begin(), tensors_infos_array.end(), + [&](const ITensorInfo *tensor_info_obj) + { return tensor_info_obj->data_layout() != tensor_data_layout; }), + function, file, line, "Tensors have different data layouts"); return arm_compute::Status{}; } /** Return an error if the passed tensors have different data layouts @@ -565,19 +610,21 @@ inline arm_compute::Status error_on_mismatching_data_layouts(const char *functio * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_data_layouts(const char *function, const char *file, const int line, - const ITensor *tensor, Ts... tensors) +inline arm_compute::Status error_on_mismatching_data_layouts( + const char *function, const char *file, const int line, const ITensor *tensor, Ts... tensors) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, std::forward<Ts>(tensors)...)); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_layouts(function, file, line, tensor->info(), - detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); + ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_layouts( + function, file, line, tensor->info(), detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_LAYOUT(...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_mismatching_data_layouts(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_mismatching_data_layouts(__func__, __FILE__, __LINE__, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_LAYOUT(...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_layouts(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_mismatching_data_layouts(__func__, __FILE__, __LINE__, __VA_ARGS__)) /** Return an error if the passed two tensor infos have different data types * @@ -590,19 +637,18 @@ inline arm_compute::Status error_on_mismatching_data_layouts(const char *functio * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_data_types(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, Ts... tensor_infos) +inline arm_compute::Status error_on_mismatching_data_types( + const char *function, const char *file, const int line, const ITensorInfo *tensor_info, Ts... tensor_infos) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, tensor_infos...)); - DataType &&tensor_data_type = tensor_info->data_type(); - const std::array<const ITensorInfo *, sizeof...(Ts)> tensors_infos_array{ { tensor_infos... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_infos_array.begin(), tensors_infos_array.end(), [&](const ITensorInfo * tensor_info_obj) - { - return tensor_info_obj->data_type() != tensor_data_type; - }), - function, file, line, "Tensors have different data types"); + DataType &&tensor_data_type = tensor_info->data_type(); + const std::array<const ITensorInfo *, sizeof...(Ts)> tensors_infos_array{{tensor_infos...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensors_infos_array.begin(), tensors_infos_array.end(), + [&](const ITensorInfo *tensor_info_obj) + { return tensor_info_obj->data_type() != tensor_data_type; }), + function, file, line, "Tensors have different data types"); return arm_compute::Status{}; } /** Return an error if the passed two tensors have different data types @@ -616,19 +662,21 @@ inline arm_compute::Status error_on_mismatching_data_types(const char *function, * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_data_types(const char *function, const char *file, const int line, - const ITensor *tensor, Ts... tensors) +inline arm_compute::Status error_on_mismatching_data_types( + const char *function, const char *file, const int line, const ITensor *tensor, Ts... tensors) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_nullptr(function, file, line, tensors...)); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_types(function, file, line, tensor->info(), - detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); + ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_types( + function, file, line, tensor->info(), detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_MISMATCHING_DATA_TYPES(...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_mismatching_data_types(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_mismatching_data_types(__func__, __FILE__, __LINE__, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_data_types(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_mismatching_data_types(__func__, __FILE__, __LINE__, __VA_ARGS__)) /** Return an error if the passed tensor infos have different asymmetric quantized data types or different quantization info * @@ -644,28 +692,32 @@ inline arm_compute::Status error_on_mismatching_data_types(const char *function, * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_quantization_info(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos) +inline arm_compute::Status error_on_mismatching_quantization_info(const char *function, + const char *file, + const int line, + const ITensorInfo *tensor_info_1, + const ITensorInfo *tensor_info_2, + Ts... tensor_infos) { DataType &&first_data_type = tensor_info_1->data_type(); const QuantizationInfo first_quantization_info = tensor_info_1->quantization_info(); - if(!is_data_type_quantized(first_data_type)) + if (!is_data_type_quantized(first_data_type)) { return arm_compute::Status{}; } - const std::array < const ITensorInfo *, 1 + sizeof...(Ts) > tensor_infos_array{ { tensor_info_2, std::forward<Ts>(tensor_infos)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensor_infos_array.begin(), tensor_infos_array.end(), [&](const ITensorInfo * tensor_info) - { - return tensor_info->data_type() != first_data_type; - }), - function, file, line, "Tensors have different asymmetric quantized data types"); - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensor_infos_array.begin(), tensor_infos_array.end(), [&](const ITensorInfo * tensor_info) - { - return tensor_info->quantization_info() != first_quantization_info; - }), - function, file, line, "Tensors have different quantization information"); + const std::array<const ITensorInfo *, 1 + sizeof...(Ts)> tensor_infos_array{ + {tensor_info_2, std::forward<Ts>(tensor_infos)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG(std::any_of(tensor_infos_array.begin(), tensor_infos_array.end(), + [&](const ITensorInfo *tensor_info) + { return tensor_info->data_type() != first_data_type; }), + function, file, line, "Tensors have different asymmetric quantized data types"); + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG( + std::any_of(tensor_infos_array.begin(), tensor_infos_array.end(), + [&](const ITensorInfo *tensor_info) + { return tensor_info->quantization_info() != first_quantization_info; }), + function, file, line, "Tensors have different quantization information"); return arm_compute::Status{}; } @@ -683,17 +735,24 @@ inline arm_compute::Status error_on_mismatching_quantization_info(const char *fu * @return Status */ template <typename... Ts> -inline arm_compute::Status error_on_mismatching_quantization_info(const char *function, const char *file, const int line, - const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors) +inline arm_compute::Status error_on_mismatching_quantization_info(const char *function, + const char *file, + const int line, + const ITensor *tensor_1, + const ITensor *tensor_2, + Ts... tensors) { - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_quantization_info(function, file, line, tensor_1->info(), tensor_2->info(), - detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); + ARM_COMPUTE_RETURN_ON_ERROR( + ::arm_compute::error_on_mismatching_quantization_info(function, file, line, tensor_1->info(), tensor_2->info(), + detail::get_tensor_info_t<ITensorInfo *>()(tensors)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_mismatching_quantization_info(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_mismatching_quantization_info(__func__, __FILE__, __LINE__, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_QUANTIZATION_INFO(...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_mismatching_quantization_info(__func__, __FILE__, __LINE__, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_mismatching_quantization_info(__func__, __FILE__, __LINE__, __VA_ARGS__)) /** Throw an error if the format of the passed tensor/multi-image does not match any of the formats provided. * @@ -705,8 +764,8 @@ inline arm_compute::Status error_on_mismatching_quantization_info(const char *fu * @param[in] formats (Optional) Further allowed formats. */ template <typename T, typename F, typename... Fs> -void error_on_format_not_in(const char *function, const char *file, const int line, - const T *object, F &&format, Fs &&... formats) +void error_on_format_not_in( + const char *function, const char *file, const int line, const T *object, F &&format, Fs &&...formats) { ARM_COMPUTE_ERROR_ON_LOC(object == nullptr, function, file, line); @@ -715,17 +774,17 @@ void error_on_format_not_in(const char *function, const char *file, const int li ARM_COMPUTE_ERROR_ON_LOC(object_format == Format::UNKNOWN, function, file, line); - const std::array<F, sizeof...(Fs)> formats_array{ { std::forward<Fs>(formats)... } }; + const std::array<F, sizeof...(Fs)> formats_array{{std::forward<Fs>(formats)...}}; ARM_COMPUTE_UNUSED(formats_array); - ARM_COMPUTE_ERROR_ON_LOC_MSG(object_format != format && std::none_of(formats_array.begin(), formats_array.end(), [&](const F & f) - { - return f == object_format; - }), - function, file, line, "Format %s not supported by this kernel", string_from_format(object_format).c_str()); + ARM_COMPUTE_ERROR_ON_LOC_MSG( + object_format != format && + std::none_of(formats_array.begin(), formats_array.end(), [&](const F &f) { return f == object_format; }), + function, file, line, "Format %s not supported by this kernel", string_from_format(object_format).c_str()); ARM_COMPUTE_UNUSED(function, format, file, line); } -#define ARM_COMPUTE_ERROR_ON_FORMAT_NOT_IN(t, ...) ::arm_compute::error_on_format_not_in(__func__, __FILE__, __LINE__, t, __VA_ARGS__) +#define ARM_COMPUTE_ERROR_ON_FORMAT_NOT_IN(t, ...) \ + ::arm_compute::error_on_format_not_in(__func__, __FILE__, __LINE__, t, __VA_ARGS__) /** Return an error if the data type of the passed tensor info does not match any of the data types provided. * @@ -739,20 +798,19 @@ void error_on_format_not_in(const char *function, const char *file, const int li * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_type_not_in(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, T &&dt, Ts &&... dts) +inline arm_compute::Status error_on_data_type_not_in( + const char *function, const char *file, const int line, const ITensorInfo *tensor_info, T &&dt, Ts &&...dts) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info == nullptr, function, file, line); const DataType &tensor_dt = tensor_info->data_type(); //NOLINT ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_dt == DataType::UNKNOWN, function, file, line); - const std::array<T, sizeof...(Ts)> dts_array{ { std::forward<Ts>(dts)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(tensor_dt != dt && std::none_of(dts_array.begin(), dts_array.end(), [&](const T & d) - { - return d == tensor_dt; - }), - function, file, line, "ITensor data type %s not supported by this kernel", string_from_data_type(tensor_dt).c_str()); + const std::array<T, sizeof...(Ts)> dts_array{{std::forward<Ts>(dts)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR( + tensor_dt != dt && std::none_of(dts_array.begin(), dts_array.end(), [&](const T &d) { return d == tensor_dt; }), + function, file, line, "ITensor data type %s not supported by this kernel", + string_from_data_type(tensor_dt).c_str()); return arm_compute::Status{}; } /** Return an error if the data type of the passed tensor does not match any of the data types provided. @@ -767,11 +825,12 @@ inline arm_compute::Status error_on_data_type_not_in(const char *function, const * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_type_not_in(const char *function, const char *file, const int line, - const ITensor *tensor, T &&dt, Ts &&... dts) +inline arm_compute::Status error_on_data_type_not_in( + const char *function, const char *file, const int line, const ITensor *tensor, T &&dt, Ts &&...dts) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_type_not_in(function, file, line, tensor->info(), std::forward<T>(dt), std::forward<Ts>(dts)...)); + ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_type_not_in( + function, file, line, tensor->info(), std::forward<T>(dt), std::forward<Ts>(dts)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_DATA_TYPE_NOT_IN(t, ...) \ @@ -791,20 +850,19 @@ inline arm_compute::Status error_on_data_type_not_in(const char *function, const * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_layout_not_in(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, T &&dl, Ts &&... dls) +inline arm_compute::Status error_on_data_layout_not_in( + const char *function, const char *file, const int line, const ITensorInfo *tensor_info, T &&dl, Ts &&...dls) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info == nullptr, function, file, line); const DataLayout &tensor_dl = tensor_info->data_layout(); //NOLINT ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_dl == DataLayout::UNKNOWN, function, file, line); - const std::array<T, sizeof...(Ts)> dls_array{ { std::forward<Ts>(dls)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(tensor_dl != dl && std::none_of(dls_array.begin(), dls_array.end(), [&](const T & l) - { - return l == tensor_dl; - }), - function, file, line, "ITensor data layout %s not supported by this kernel", string_from_data_layout(tensor_dl).c_str()); + const std::array<T, sizeof...(Ts)> dls_array{{std::forward<Ts>(dls)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR( + tensor_dl != dl && std::none_of(dls_array.begin(), dls_array.end(), [&](const T &l) { return l == tensor_dl; }), + function, file, line, "ITensor data layout %s not supported by this kernel", + string_from_data_layout(tensor_dl).c_str()); return arm_compute::Status{}; } /** Return an error if the data layout of the passed tensor does not match any of the data layout provided. @@ -819,17 +877,19 @@ inline arm_compute::Status error_on_data_layout_not_in(const char *function, con * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_layout_not_in(const char *function, const char *file, const int line, - const ITensor *tensor, T &&dl, Ts &&... dls) +inline arm_compute::Status error_on_data_layout_not_in( + const char *function, const char *file, const int line, const ITensor *tensor, T &&dl, Ts &&...dls) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_layout_not_in(function, file, line, tensor->info(), std::forward<T>(dl), std::forward<Ts>(dls)...)); + ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_layout_not_in( + function, file, line, tensor->info(), std::forward<T>(dl), std::forward<Ts>(dls)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_DATA_LAYOUT_NOT_IN(t, ...) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_data_layout_not_in(__func__, __FILE__, __LINE__, t, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_DATA_LAYOUT_NOT_IN(t, ...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_layout_not_in(__func__, __FILE__, __LINE__, t, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_data_layout_not_in(__func__, __FILE__, __LINE__, t, __VA_ARGS__)) /** Return an error if the data type or the number of channels of the passed tensor info does not match any of the data types and number of channels provided. * @@ -844,12 +904,20 @@ inline arm_compute::Status error_on_data_layout_not_in(const char *function, con * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_type_channel_not_in(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, size_t num_channels, T &&dt, Ts &&... dts) +inline arm_compute::Status error_on_data_type_channel_not_in(const char *function, + const char *file, + const int line, + const ITensorInfo *tensor_info, + size_t num_channels, + T &&dt, + Ts &&...dts) { - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_type_not_in(function, file, line, tensor_info, std::forward<T>(dt), std::forward<Ts>(dts)...)); + ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_type_not_in( + function, file, line, tensor_info, std::forward<T>(dt), std::forward<Ts>(dts)...)); const size_t tensor_nc = tensor_info->num_channels(); - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(tensor_nc != num_channels, function, file, line, "Number of channels %zu. Required number of channels %zu", tensor_nc, num_channels); + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG_VAR(tensor_nc != num_channels, function, file, line, + "Number of channels %zu. Required number of channels %zu", tensor_nc, + num_channels); return arm_compute::Status{}; } /** Return an error if the data type or the number of channels of the passed tensor does not match any of the data types and number of channels provided. @@ -865,17 +933,25 @@ inline arm_compute::Status error_on_data_type_channel_not_in(const char *functio * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_data_type_channel_not_in(const char *function, const char *file, const int line, - const ITensor *tensor, size_t num_channels, T &&dt, Ts &&... dts) +inline arm_compute::Status error_on_data_type_channel_not_in(const char *function, + const char *file, + const int line, + const ITensor *tensor, + size_t num_channels, + T &&dt, + Ts &&...dts) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); - ARM_COMPUTE_RETURN_ON_ERROR(error_on_data_type_channel_not_in(function, file, line, tensor->info(), num_channels, std::forward<T>(dt), std::forward<Ts>(dts)...)); + ARM_COMPUTE_RETURN_ON_ERROR(error_on_data_type_channel_not_in(function, file, line, tensor->info(), num_channels, + std::forward<T>(dt), std::forward<Ts>(dts)...)); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c, ...) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_data_type_channel_not_in(__func__, __FILE__, __LINE__, t, c, __VA_ARGS__)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_data_type_channel_not_in(__func__, __FILE__, __LINE__, t, c, __VA_ARGS__)) #define ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(t, c, ...) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_data_type_channel_not_in(__func__, __FILE__, __LINE__, t, c, __VA_ARGS__)) + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_data_type_channel_not_in(__func__, __FILE__, __LINE__, t, c, __VA_ARGS__)) /** Return an error if the data type of the passed tensor info is FP16 and FP16 extension is not supported by the device. * @@ -887,12 +963,12 @@ inline arm_compute::Status error_on_data_type_channel_not_in(const char *functio * * @return Status */ -inline arm_compute::Status error_on_unsupported_fp16(const char *function, const char *file, const int line, - const ITensorInfo *tensor_info, bool is_fp16_supported) +inline arm_compute::Status error_on_unsupported_fp16( + const char *function, const char *file, const int line, const ITensorInfo *tensor_info, bool is_fp16_supported) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor_info == nullptr, function, file, line); - ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG((tensor_info->data_type() == DataType::F16 && !is_fp16_supported), - function, file, line, "FP16 not supported by the device"); + ARM_COMPUTE_RETURN_ERROR_ON_LOC_MSG((tensor_info->data_type() == DataType::F16 && !is_fp16_supported), function, + file, line, "FP16 not supported by the device"); return arm_compute::Status{}; } @@ -906,11 +982,12 @@ inline arm_compute::Status error_on_unsupported_fp16(const char *function, const * * @return Status */ -inline arm_compute::Status error_on_unsupported_fp16(const char *function, const char *file, const int line, - const ITensor *tensor, bool is_fp16_supported) +inline arm_compute::Status error_on_unsupported_fp16( + const char *function, const char *file, const int line, const ITensor *tensor, bool is_fp16_supported) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(tensor == nullptr, function, file, line); - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_unsupported_fp16(function, file, line, tensor->info(), is_fp16_supported)); + ARM_COMPUTE_RETURN_ON_ERROR( + ::arm_compute::error_on_unsupported_fp16(function, file, line, tensor->info(), is_fp16_supported)); return arm_compute::Status{}; } @@ -923,8 +1000,8 @@ inline arm_compute::Status error_on_unsupported_fp16(const char *function, const * * @return Status */ -arm_compute::Status error_on_tensor_not_2d(const char *function, const char *file, const int line, - const ITensor *tensor); +arm_compute::Status +error_on_tensor_not_2d(const char *function, const char *file, const int line, const ITensor *tensor); /** Return an error if the tensor info is not 2D. * @@ -935,8 +1012,8 @@ arm_compute::Status error_on_tensor_not_2d(const char *function, const char *fil * * @return Status */ -arm_compute::Status error_on_tensor_not_2d(const char *function, const char *file, const int line, - const ITensorInfo *tensor); +arm_compute::Status +error_on_tensor_not_2d(const char *function, const char *file, const int line, const ITensorInfo *tensor); #define ARM_COMPUTE_ERROR_ON_TENSOR_NOT_2D(t) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_tensor_not_2d(__func__, __FILE__, __LINE__, t)) @@ -955,17 +1032,15 @@ arm_compute::Status error_on_tensor_not_2d(const char *function, const char *fil * @return Status */ template <typename T, typename... Ts> -inline arm_compute::Status error_on_channel_not_in(const char *function, const char *file, const int line, - T cn, T &&channel, Ts &&... channels) +inline arm_compute::Status +error_on_channel_not_in(const char *function, const char *file, const int line, T cn, T &&channel, Ts &&...channels) { ARM_COMPUTE_RETURN_ERROR_ON_LOC(cn == Channel::UNKNOWN, function, file, line); - const std::array<T, sizeof...(Ts)> channels_array{ { std::forward<Ts>(channels)... } }; - ARM_COMPUTE_RETURN_ERROR_ON_LOC(channel != cn && std::none_of(channels_array.begin(), channels_array.end(), [&](const T & f) - { - return f == cn; - }), - function, file, line); + const std::array<T, sizeof...(Ts)> channels_array{{std::forward<Ts>(channels)...}}; + ARM_COMPUTE_RETURN_ERROR_ON_LOC(channel != cn && std::none_of(channels_array.begin(), channels_array.end(), + [&](const T &f) { return f == cn; }), + function, file, line); return arm_compute::Status{}; } #define ARM_COMPUTE_ERROR_ON_CHANNEL_NOT_IN(c, ...) \ @@ -983,8 +1058,8 @@ inline arm_compute::Status error_on_channel_not_in(const char *function, const c * * @return Status */ -arm_compute::Status error_on_channel_not_in_known_format(const char *function, const char *file, const int line, - Format fmt, Channel cn); +arm_compute::Status +error_on_channel_not_in_known_format(const char *function, const char *file, const int line, Format fmt, Channel cn); #define ARM_COMPUTE_ERROR_ON_CHANNEL_NOT_IN_KNOWN_FORMAT(f, c) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_channel_not_in_known_format(__func__, __FILE__, __LINE__, f, c)) #define ARM_COMPUTE_RETURN_ERROR_ON_CHANNEL_NOT_IN_KNOWN_FORMAT(f, c) \ @@ -999,8 +1074,8 @@ arm_compute::Status error_on_channel_not_in_known_format(const char *function, c * * @return Status */ -arm_compute::Status error_on_unconfigured_kernel(const char *function, const char *file, const int line, - const IKernel *kernel); +arm_compute::Status +error_on_unconfigured_kernel(const char *function, const char *file, const int line, const IKernel *kernel); #define ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(k) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_unconfigured_kernel(__func__, __FILE__, __LINE__, k)) #define ARM_COMPUTE_RETURN_ERROR_ON_UNCONFIGURED_KERNEL(k) \ @@ -1017,8 +1092,12 @@ arm_compute::Status error_on_unconfigured_kernel(const char *function, const cha * * @return Status */ -arm_compute::Status error_on_invalid_subtensor(const char *function, const char *file, const int line, - const TensorShape &parent_shape, const Coordinates &coords, const TensorShape &shape); +arm_compute::Status error_on_invalid_subtensor(const char *function, + const char *file, + const int line, + const TensorShape &parent_shape, + const Coordinates &coords, + const TensorShape &shape); #define ARM_COMPUTE_ERROR_ON_INVALID_SUBTENSOR(p, c, s) \ ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_invalid_subtensor(__func__, __FILE__, __LINE__, p, c, s)) #define ARM_COMPUTE_RETURN_ERROR_ON_INVALID_SUBTENSOR(p, c, s) \ @@ -1034,11 +1113,16 @@ arm_compute::Status error_on_invalid_subtensor(const char *function, const char * * @return Status */ -arm_compute::Status error_on_invalid_subtensor_valid_region(const char *function, const char *file, const int line, - const ValidRegion &parent_valid_region, const ValidRegion &valid_region); +arm_compute::Status error_on_invalid_subtensor_valid_region(const char *function, + const char *file, + const int line, + const ValidRegion &parent_valid_region, + const ValidRegion &valid_region); #define ARM_COMPUTE_ERROR_ON_INVALID_SUBTENSOR_VALID_REGION(pv, sv) \ - ARM_COMPUTE_ERROR_THROW_ON(::arm_compute::error_on_invalid_subtensor_valid_region(__func__, __FILE__, __LINE__, pv, sv)) + ARM_COMPUTE_ERROR_THROW_ON( \ + ::arm_compute::error_on_invalid_subtensor_valid_region(__func__, __FILE__, __LINE__, pv, sv)) #define ARM_COMPUTE_RETURN_ERROR_ON_INVALID_SUBTENSOR_VALID_REGION(pv, sv) \ - ARM_COMPUTE_RETURN_ON_ERROR(::arm_compute::error_on_invalid_subtensor_valid_region(__func__, __FILE__, __LINE__, pv, sv)) -} + ARM_COMPUTE_RETURN_ON_ERROR( \ + ::arm_compute::error_on_invalid_subtensor_valid_region(__func__, __FILE__, __LINE__, pv, sv)) +} // namespace arm_compute #endif /* ARM_COMPUTE_VALIDATE_H*/ diff --git a/arm_compute/core/Version.h b/arm_compute/core/Version.h index a4d307950a..44d400bad8 100644 --- a/arm_compute/core/Version.h +++ b/arm_compute/core/Version.h @@ -28,7 +28,7 @@ /* Macro utilities */ #define ARM_COMPUTE_STRINGIFY2(s) #s -#define ARM_COMPUTE_STRINGIFY(s) ARM_COMPUTE_STRINGIFY2(s) +#define ARM_COMPUTE_STRINGIFY(s) ARM_COMPUTE_STRINGIFY2(s) #define ARM_COMPUTE_VERSION_STR \ ARM_COMPUTE_STRINGIFY(ARM_COMPUTE_VERSION_MAJOR) \ diff --git a/arm_compute/core/Window.h b/arm_compute/core/Window.h index 8ae859f4b3..4863b95045 100644 --- a/arm_compute/core/Window.h +++ b/arm_compute/core/Window.h @@ -24,15 +24,15 @@ #ifndef ARM_COMPUTE_WINDOW_H #define ARM_COMPUTE_WINDOW_H -#include <algorithm> -#include <array> -#include <cstddef> - #include "arm_compute/core/Coordinates.h" #include "arm_compute/core/Error.h" #include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/utils/math/Math.h" +#include <algorithm> +#include <array> +#include <cstddef> + namespace arm_compute { /** Describe a multidimensional execution window. */ @@ -86,8 +86,7 @@ public: * @param[in] step Step between two elements of the dimension when iterating. * */ - constexpr Dimension(int start = 0, int end = 1, int step = 1) - : _start(start), _end(end), _step(step) + constexpr Dimension(int start = 0, int end = 1, int step = 1) : _start(start), _end(end), _step(step) { } Dimension(const Dimension &d) = default; @@ -373,7 +372,8 @@ public: * * @return Collapsed window. */ - Window collapse_if_possible(const Window &full_window, size_t first, size_t last, bool *has_collapsed = nullptr) const; + Window + collapse_if_possible(const Window &full_window, size_t first, size_t last, bool *has_collapsed = nullptr) const; /** Collapse the dimensions higher than @p first if possible. * @@ -441,7 +441,7 @@ private: * @return The first slice of the window. */ template <unsigned int window_dimension> - Window first_slice_window() const; + Window first_slice_window() const; /** Slide the passed window slice. * diff --git a/arm_compute/core/Window.inl b/arm_compute/core/Window.inl index 5ee4b57145..d935507b1d 100644 --- a/arm_compute/core/Window.inl +++ b/arm_compute/core/Window.inl @@ -26,7 +26,7 @@ namespace arm_compute inline Window::Window(const Window &src) : _dims(), _is_broadcasted(utility::generate_array<bool, Coordinates::num_max_dimensions, false>::value) { - for(size_t i = 0; i < Coordinates::num_max_dimensions; ++i) + for (size_t i = 0; i < Coordinates::num_max_dimensions; ++i) { set(i, src[i]); _is_broadcasted[i] = src.is_broadcasted(i); @@ -65,32 +65,34 @@ inline bool Window::is_broadcasted(size_t dimension) const return _is_broadcasted[dimension]; } -inline Window Window::collapse_if_possible(const Window &full_window, const size_t first, - const size_t last, bool *has_collapsed) const +inline Window Window::collapse_if_possible(const Window &full_window, + const size_t first, + const size_t last, + bool *has_collapsed) const { Window collapsed(*this); bool is_collapsable = true; int collapsed_end = _dims[first].end(); - for(size_t d = first + 1; is_collapsable && (d < last); ++d) + for (size_t d = first + 1; is_collapsable && (d < last); ++d) { // The _dims's dimension must match the full _dims dimension to be collapsable: - is_collapsable = (_dims[d].start() == 0) && (full_window[d].start() == 0) && (_dims[d].step() <= 1) - && (full_window[d].end() == _dims[d].end()); + is_collapsable = (_dims[d].start() == 0) && (full_window[d].start() == 0) && (_dims[d].step() <= 1) && + (full_window[d].end() == _dims[d].end()); collapsed_end *= _dims[d].end(); } - if(is_collapsable) + if (is_collapsable) { collapsed._dims.at(first).set_end(collapsed_end); - for(size_t d = first + 1; is_collapsable && (d < last); ++d) + for (size_t d = first + 1; is_collapsable && (d < last); ++d) { collapsed.set(d, Dimension()); } } - if(has_collapsed != nullptr) + if (has_collapsed != nullptr) { *has_collapsed = is_collapsable; } @@ -101,7 +103,7 @@ inline Window Window::collapse_if_possible(const Window &full_window, const size inline Window Window::shift_dimensions(unsigned int shift_value) const { Window shifted_window; - for(size_t n = 0; n < (Coordinates::num_max_dimensions - shift_value); n++) + for (size_t n = 0; n < (Coordinates::num_max_dimensions - shift_value); n++) { shifted_window.set(n, _dims[n + shift_value]); } @@ -120,9 +122,9 @@ inline Window Window::collapse(const Window &full_window, const size_t first, co inline Window Window::broadcast_if_dimension_le_one(const TensorShape &shape) const { Window broadcastWin(*this); - for(size_t d = 0; d < TensorShape::num_max_dimensions; ++d) + for (size_t d = 0; d < TensorShape::num_max_dimensions; ++d) { - if(shape[d] <= 1) + if (shape[d] <= 1) { broadcastWin.set_broadcasted(d); } @@ -142,7 +144,7 @@ inline void Window::adjust(size_t dimension, int adjust_value, bool is_at_start) ARM_COMPUTE_ERROR_ON(dimension >= Coordinates::num_max_dimensions); Window::Dimension &d = _dims[dimension]; - if(is_at_start) + if (is_at_start) { d = Window::Dimension(d.start() + adjust_value, d.end(), d.step()); } @@ -172,7 +174,7 @@ inline void Window::set_dimension_step(size_t dimension, int step) inline void Window::validate() const { - for(size_t i = 0; i < Coordinates::num_max_dimensions; ++i) + for (size_t i = 0; i < Coordinates::num_max_dimensions; ++i) { ARM_COMPUTE_ERROR_ON(_dims[i].end() < _dims[i].start()); ARM_COMPUTE_ERROR_ON((_dims[i].step() != 0) && (((_dims[i].end() - _dims[i].start()) % _dims[i].step()) != 0)); @@ -193,9 +195,9 @@ inline Window Window::split_window(size_t dimension, size_t id, size_t total) co Window out; - for(size_t d = 0; d < Coordinates::num_max_dimensions; ++d) + for (size_t d = 0; d < Coordinates::num_max_dimensions; ++d) { - if(d == dimension) + if (d == dimension) { int start = _dims[d].start(); int end = _dims[d].end(); @@ -207,7 +209,7 @@ inline Window Window::split_window(size_t dimension, size_t id, size_t total) co int it_start = work * id; - if(int(id) < rem) + if (int(id) < rem) { ++work; it_start += id; @@ -234,18 +236,18 @@ inline Window Window::split_window(size_t dimension, size_t id, size_t total) co template <unsigned int window_dimension> inline bool Window::slide_window_slice(Window &slice) const { - for(unsigned int n = window_dimension; n < Coordinates::num_max_dimensions; ++n) + for (unsigned int n = window_dimension; n < Coordinates::num_max_dimensions; ++n) { // Did we reach the end of this dimension? const int v = slice._dims[n].start() + 1; - if(v < _dims[n].end()) + if (v < _dims[n].end()) { // No: increment slice._dims[n] = Dimension(v, v + 1, 1); // Reset lower dimensions: - for(unsigned int lower = window_dimension; lower < n; ++lower) + for (unsigned int lower = window_dimension; lower < n; ++lower) { slice._dims[lower] = Dimension(_dims[lower].start(), _dims[lower].start() + 1, 1); } @@ -258,14 +260,14 @@ inline bool Window::slide_window_slice(Window &slice) const } template <unsigned int window_dimension> -inline Window Window::first_slice_window() const +inline Window Window::first_slice_window() const { Window slice; std::copy_n(_dims.begin(), window_dimension, slice._dims.begin()); //Initialise higher dimensions to be the first slice. - for(unsigned int n = window_dimension; n < Coordinates::num_max_dimensions; ++n) + for (unsigned int n = window_dimension; n < Coordinates::num_max_dimensions; ++n) { slice._dims[n] = Dimension(_dims[n].start(), _dims[n].start() + 1, 1); } @@ -275,7 +277,7 @@ inline Window Window::first_slice_window() const inline void Window::use_tensor_dimensions(const TensorShape &shape, size_t first_dimension) { - for(unsigned int n = first_dimension; n < shape.num_dimensions(); ++n) + for (unsigned int n = first_dimension; n < shape.num_dimensions(); ++n) { set(n, Window::Dimension(0, std::max(shape[n], static_cast<size_t>(1)))); } @@ -284,7 +286,7 @@ inline void Window::use_tensor_dimensions(const TensorShape &shape, size_t first inline TensorShape Window::shape() const { TensorShape shape; - for(size_t d = 0; d < TensorShape::num_max_dimensions; ++d) + for (size_t d = 0; d < TensorShape::num_max_dimensions; ++d) { shape.set(d, (_dims[d].end() - _dims[d].start()) / _dims[d].step()); } @@ -294,7 +296,7 @@ inline TensorShape Window::shape() const inline size_t Window::num_iterations_total() const { size_t total = 1; - for(size_t d = 0; d < Coordinates::num_max_dimensions; ++d) + for (size_t d = 0; d < Coordinates::num_max_dimensions; ++d) { total *= num_iterations(d); } diff --git a/arm_compute/core/WindowIterator.h b/arm_compute/core/WindowIterator.h index b1e399c872..29302c410a 100644 --- a/arm_compute/core/WindowIterator.h +++ b/arm_compute/core/WindowIterator.h @@ -28,7 +28,6 @@ #include "arm_compute/core/ITensor.h" #include "arm_compute/core/Window.h" - namespace arm_compute { /** Convert an offset in window steps into absolute coordinates. @@ -41,7 +40,7 @@ namespace arm_compute inline Coordinates convert_window_coord_to_position(const Window &w, const Coordinates &offset) { Coordinates position; - for(unsigned int i = 0; i < Coordinates::num_max_dimensions; ++i) + for (unsigned int i = 0; i < Coordinates::num_max_dimensions; ++i) { position.set(i, w[i].start() + offset[i] * w[i].step()); } @@ -165,7 +164,7 @@ public: template <typename M> void iterate_3D(M &&on_new_row_size) { - while(_end.z() != _position.z()) + while (_end.z() != _position.z()) { iterate_2D_internal(on_new_row_size, _w.x().end() - _w.x().step(), _w.y().end() - _w.y().step()); _position[2] += _w.z().step(); @@ -212,7 +211,7 @@ private: void iterate_2D_internal(M &&on_new_row_size, int end_x, int end_y) { //Is there more than one row to process ? - if(end_y == _position.y()) + if (end_y == _position.y()) { // Both start and end belong to the same row: iterate_over_dim0(end_x + _w.x().step(), on_new_row_size); @@ -220,7 +219,7 @@ private: else { // Do we start from the beginning of the row ? - if(_w.x().start() != _position.x()) + if (_w.x().start() != _position.x()) { //Start in the middle of a row: process left-over X iterate_over_dim0(_w.x().end(), on_new_row_size); @@ -229,7 +228,7 @@ private: //Middle rows bool no_leftover = end_x + _w.x().step() == _w.x().end(); - if(no_leftover) + if (no_leftover) { //Switch to full row size: on_new_row_size(_w[0].start(), _w.x().end()); @@ -241,7 +240,7 @@ private: else { // Are there full rows to process ? - if(_position[1] != end_y) + if (_position[1] != end_y) { //Switch to full row size: on_new_row_size(_w[0].start(), _w.x().end()); @@ -261,7 +260,7 @@ private: */ void iterate_over_dim1(int end) { - for(; _position[1] != end; _position[1] += _w[1].step()) + for (; _position[1] != end; _position[1] += _w[1].step()) { _position[0] = _w[0].start(); iterate_over_dim0(_w[0].end()); @@ -288,7 +287,7 @@ private: { // Both start and end belong to the same row: ARM_COMPUTE_ERROR_ON(_position[0] > end); - for(; _position.x() < end; _position[0] += _w[0].step()) + for (; _position.x() < end; _position[0] += _w[0].step()) { _lambda_function(_position); } @@ -310,9 +309,10 @@ private: * @return A WindowIterator object. */ template <typename L> -WindowIterator<L> create_window_iterator(const Window &w, const Coordinates &start, const Coordinates &end, L &&lambda_function) +WindowIterator<L> +create_window_iterator(const Window &w, const Coordinates &start, const Coordinates &end, L &&lambda_function) { return WindowIterator<L>(w, start, end, std::move(lambda_function)); } -} +} // namespace arm_compute #endif /*ARM_COMPUTE_WINDOW_ITERATOR_H*/ diff --git a/arm_compute/core/experimental/Types.h b/arm_compute/core/experimental/Types.h index 8dd6812b58..63a3a1a1ec 100644 --- a/arm_compute/core/experimental/Types.h +++ b/arm_compute/core/experimental/Types.h @@ -92,24 +92,18 @@ struct MemoryInfo { MemoryInfo() = default; - MemoryInfo(int slot, size_t size, size_t alignment = 0) noexcept - : slot(slot), - size(size), - alignment(alignment) + MemoryInfo(int slot, size_t size, size_t alignment = 0) noexcept : slot(slot), size(size), alignment(alignment) { } MemoryInfo(int slot, MemoryLifetime lifetime, size_t size, size_t alignment = 0) noexcept - : slot(slot), - lifetime(lifetime), - size(size), - alignment(alignment) + : slot(slot), lifetime(lifetime), size(size), alignment(alignment) { } bool merge(int slot, size_t new_size, size_t new_alignment = 0) noexcept { - if(slot != this->slot) + if (slot != this->slot) { return false; } @@ -120,10 +114,10 @@ struct MemoryInfo return true; } - int slot{ ACL_UNKNOWN }; - MemoryLifetime lifetime{ MemoryLifetime::Temporary }; - size_t size{ 0 }; - size_t alignment{ 64 }; + int slot{ACL_UNKNOWN}; + MemoryLifetime lifetime{MemoryLifetime::Temporary}; + size_t size{0}; + size_t alignment{64}; }; using MemoryRequirements = std::vector<MemoryInfo>; diff --git a/arm_compute/core/utils/ActivationFunctionUtils.h b/arm_compute/core/utils/ActivationFunctionUtils.h index 1cb66da13d..c988efa256 100644 --- a/arm_compute/core/utils/ActivationFunctionUtils.h +++ b/arm_compute/core/utils/ActivationFunctionUtils.h @@ -37,5 +37,5 @@ namespace arm_compute * @return The string describing the activation function. */ const std::string &string_from_activation_func(const ActivationFunction &act); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_ACTIVATIONFUNCTIONUTILS_H */ diff --git a/arm_compute/core/utils/DataLayoutUtils.h b/arm_compute/core/utils/DataLayoutUtils.h index 399f55c63f..61839c9f91 100644 --- a/arm_compute/core/utils/DataLayoutUtils.h +++ b/arm_compute/core/utils/DataLayoutUtils.h @@ -36,5 +36,5 @@ namespace arm_compute * @return The string describing the data layout. */ const std::string &string_from_data_layout(DataLayout dl); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_DATALAYOUTUTILS_H */ diff --git a/arm_compute/core/utils/DataTypeUtils.h b/arm_compute/core/utils/DataTypeUtils.h index cbb409c8a1..7ea5eb7670 100644 --- a/arm_compute/core/utils/DataTypeUtils.h +++ b/arm_compute/core/utils/DataTypeUtils.h @@ -37,7 +37,7 @@ namespace arm_compute */ inline size_t data_size_from_type(DataType data_type) { - switch(data_type) + switch (data_type) { case DataType::U8: case DataType::S8: @@ -77,7 +77,7 @@ inline size_t data_size_from_type(DataType data_type) */ inline size_t element_size_from_data_type(DataType dt) { - switch(dt) + switch (dt) { case DataType::S8: case DataType::U8: @@ -114,7 +114,7 @@ inline size_t element_size_from_data_type(DataType dt) */ inline DataType data_type_from_format(Format format) { - switch(format) + switch (format) { case Format::U8: case Format::UV88: @@ -158,7 +158,7 @@ inline DataType data_type_from_format(Format format) */ inline DataType get_promoted_data_type(DataType dt) { - switch(dt) + switch (dt) { case DataType::U8: return DataType::U16; @@ -196,7 +196,7 @@ inline std::tuple<PixelValue, PixelValue> get_min_max(DataType dt) { PixelValue min{}; PixelValue max{}; - switch(dt) + switch (dt) { case DataType::U8: case DataType::QASYMM8: @@ -303,7 +303,7 @@ inline ::std::istream &operator>>(::std::istream &stream, DataType &data_type) */ inline bool is_data_type_float(DataType dt) { - switch(dt) + switch (dt) { case DataType::F16: case DataType::F32: @@ -323,7 +323,7 @@ inline bool is_data_type_float(DataType dt) */ inline bool is_data_type_quantized(DataType dt) { - switch(dt) + switch (dt) { case DataType::QSYMM8: case DataType::QASYMM8: @@ -345,7 +345,7 @@ inline bool is_data_type_quantized(DataType dt) */ inline bool is_data_type_quantized_asymmetric(DataType dt) { - switch(dt) + switch (dt) { case DataType::QASYMM8: case DataType::QASYMM8_SIGNED: @@ -364,7 +364,7 @@ inline bool is_data_type_quantized_asymmetric(DataType dt) */ inline bool is_data_type_quantized_asymmetric_signed(DataType dt) { - switch(dt) + switch (dt) { case DataType::QASYMM8_SIGNED: return true; @@ -381,7 +381,7 @@ inline bool is_data_type_quantized_asymmetric_signed(DataType dt) */ inline bool is_data_type_quantized_symmetric(DataType dt) { - switch(dt) + switch (dt) { case DataType::QSYMM8: case DataType::QSYMM8_PER_CHANNEL: @@ -400,7 +400,7 @@ inline bool is_data_type_quantized_symmetric(DataType dt) */ inline bool is_data_type_quantized_per_channel(DataType dt) { - switch(dt) + switch (dt) { case DataType::QSYMM8_PER_CHANNEL: return true; @@ -420,12 +420,13 @@ inline bool is_data_type_quantized_per_channel(DataType dt) template <typename T> bool check_value_range(T val, DataType dt, QuantizationInfo qinfo = QuantizationInfo()) { - switch(dt) + switch (dt) { case DataType::U8: { const auto val_u8 = static_cast<uint8_t>(val); - return ((val_u8 == val) && val >= std::numeric_limits<uint8_t>::lowest() && val <= std::numeric_limits<uint8_t>::max()); + return ((val_u8 == val) && val >= std::numeric_limits<uint8_t>::lowest() && + val <= std::numeric_limits<uint8_t>::max()); } case DataType::QASYMM8: { @@ -436,29 +437,34 @@ bool check_value_range(T val, DataType dt, QuantizationInfo qinfo = Quantization case DataType::S8: { const auto val_s8 = static_cast<int8_t>(val); - return ((val_s8 == val) && val >= std::numeric_limits<int8_t>::lowest() && val <= std::numeric_limits<int8_t>::max()); + return ((val_s8 == val) && val >= std::numeric_limits<int8_t>::lowest() && + val <= std::numeric_limits<int8_t>::max()); } case DataType::U16: { const auto val_u16 = static_cast<uint16_t>(val); - return ((val_u16 == val) && val >= std::numeric_limits<uint16_t>::lowest() && val <= std::numeric_limits<uint16_t>::max()); + return ((val_u16 == val) && val >= std::numeric_limits<uint16_t>::lowest() && + val <= std::numeric_limits<uint16_t>::max()); } case DataType::S16: { const auto val_s16 = static_cast<int16_t>(val); - return ((val_s16 == val) && val >= std::numeric_limits<int16_t>::lowest() && val <= std::numeric_limits<int16_t>::max()); + return ((val_s16 == val) && val >= std::numeric_limits<int16_t>::lowest() && + val <= std::numeric_limits<int16_t>::max()); } case DataType::U32: { const auto val_d64 = static_cast<double>(val); const auto val_u32 = static_cast<uint32_t>(val); - return ((val_u32 == val_d64) && val_d64 >= std::numeric_limits<uint32_t>::lowest() && val_d64 <= std::numeric_limits<uint32_t>::max()); + return ((val_u32 == val_d64) && val_d64 >= std::numeric_limits<uint32_t>::lowest() && + val_d64 <= std::numeric_limits<uint32_t>::max()); } case DataType::S32: { const auto val_d64 = static_cast<double>(val); const auto val_s32 = static_cast<int32_t>(val); - return ((val_s32 == val_d64) && val_d64 >= std::numeric_limits<int32_t>::lowest() && val_d64 <= std::numeric_limits<int32_t>::max()); + return ((val_s32 == val_d64) && val_d64 >= std::numeric_limits<int32_t>::lowest() && + val_d64 <= std::numeric_limits<int32_t>::max()); } case DataType::BFLOAT16: return (val >= bfloat16::lowest() && val <= bfloat16::max()); @@ -482,7 +488,7 @@ inline std::string cpu_impl_dt(const DataType &data_type) { std::string ret = ""; - switch(data_type) + switch (data_type) { case DataType::F32: ret = "fp32"; @@ -521,5 +527,5 @@ inline std::string cpu_impl_dt(const DataType &data_type) return ret; } -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_DATATYPEUTILS_H */ diff --git a/arm_compute/core/utils/FormatUtils.h b/arm_compute/core/utils/FormatUtils.h index afb0f78255..a8e96bd361 100644 --- a/arm_compute/core/utils/FormatUtils.h +++ b/arm_compute/core/utils/FormatUtils.h @@ -37,7 +37,7 @@ namespace arm_compute */ inline size_t pixel_size_from_format(Format format) { - switch(format) + switch (format) { case Format::U8: return 1; @@ -77,7 +77,7 @@ inline size_t pixel_size_from_format(Format format) */ inline int plane_idx_from_channel(Format format, Channel channel) { - switch(format) + switch (format) { // Single planar formats have a single plane case Format::U8: @@ -99,7 +99,7 @@ inline int plane_idx_from_channel(Format format, Channel channel) case Format::NV21: { // Channel U and V share the same plane of format UV88 - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -114,7 +114,7 @@ inline int plane_idx_from_channel(Format format, Channel channel) case Format::IYUV: case Format::YUV444: { - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -142,11 +142,11 @@ inline int plane_idx_from_channel(Format format, Channel channel) */ inline int channel_idx_from_format(Format format, Channel channel) { - switch(format) + switch (format) { case Format::RGB888: { - switch(channel) + switch (channel) { case Channel::R: return 0; @@ -161,7 +161,7 @@ inline int channel_idx_from_format(Format format, Channel channel) } case Format::RGBA8888: { - switch(channel) + switch (channel) { case Channel::R: return 0; @@ -178,7 +178,7 @@ inline int channel_idx_from_format(Format format, Channel channel) } case Format::YUYV422: { - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -193,7 +193,7 @@ inline int channel_idx_from_format(Format format, Channel channel) } case Format::UYVY422: { - switch(channel) + switch (channel) { case Channel::Y: return 1; @@ -208,7 +208,7 @@ inline int channel_idx_from_format(Format format, Channel channel) } case Format::NV12: { - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -223,7 +223,7 @@ inline int channel_idx_from_format(Format format, Channel channel) } case Format::NV21: { - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -239,7 +239,7 @@ inline int channel_idx_from_format(Format format, Channel channel) case Format::YUV444: case Format::IYUV: { - switch(channel) + switch (channel) { case Channel::Y: return 0; @@ -266,7 +266,7 @@ inline int channel_idx_from_format(Format format, Channel channel) */ inline size_t num_planes_from_format(Format format) { - switch(format) + switch (format) { case Format::U8: case Format::S16: @@ -301,7 +301,7 @@ inline size_t num_planes_from_format(Format format) */ inline size_t num_channels_from_format(Format format) { - switch(format) + switch (format) { case Format::U8: case Format::U16: @@ -340,5 +340,5 @@ inline size_t num_channels_from_format(Format format) * @return The string describing the format. */ const std::string &string_from_format(Format format); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_FORMATUTILS_H */ diff --git a/arm_compute/core/utils/InterpolationPolicyUtils.h b/arm_compute/core/utils/InterpolationPolicyUtils.h index 79f6e3aa5f..8d4ae4321c 100644 --- a/arm_compute/core/utils/InterpolationPolicyUtils.h +++ b/arm_compute/core/utils/InterpolationPolicyUtils.h @@ -37,5 +37,5 @@ namespace arm_compute * @return The string describing the interpolation policy. */ const std::string &string_from_interpolation_policy(InterpolationPolicy policy); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_INTERPOLATIONPOLICYUTILS_H */ diff --git a/arm_compute/core/utils/StringUtils.h b/arm_compute/core/utils/StringUtils.h index 41f29b0901..c13cbaa334 100644 --- a/arm_compute/core/utils/StringUtils.h +++ b/arm_compute/core/utils/StringUtils.h @@ -61,5 +61,5 @@ std::string float_to_string_with_full_precision(float val); * @return std::string */ std::string join(const std::vector<std::string> strings, const std::string &sep); -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CORE_UTILS_STRINGUTILS_H */ diff --git a/arm_compute/core/utils/helpers/AdjustVecSize.h b/arm_compute/core/utils/helpers/AdjustVecSize.h index bbb3048b84..842e3b57d6 100644 --- a/arm_compute/core/utils/helpers/AdjustVecSize.h +++ b/arm_compute/core/utils/helpers/AdjustVecSize.h @@ -39,17 +39,17 @@ inline unsigned int adjust_vec_size(unsigned int vec_size, size_t dim0) { ARM_COMPUTE_ERROR_ON(vec_size > 16); - if((vec_size >= dim0) && (dim0 == 3)) + if ((vec_size >= dim0) && (dim0 == 3)) { return dim0; } - while(vec_size > dim0) + while (vec_size > dim0) { vec_size >>= 1; } return vec_size; } -} +} // namespace arm_compute #endif /*ARM_COMPUTE_UTILS_H */ diff --git a/arm_compute/core/utils/helpers/tensor_transform.h b/arm_compute/core/utils/helpers/tensor_transform.h index faa5b4433c..7a61fa192a 100644 --- a/arm_compute/core/utils/helpers/tensor_transform.h +++ b/arm_compute/core/utils/helpers/tensor_transform.h @@ -52,7 +52,8 @@ int calculate_stride_on_index(int index, Coordinates strides); * * @return Absolute start position of a given index */ -int calculate_start_on_index(TensorShape input_shape, int index, Coordinates starts, Coordinates strides, int32_t begin_mask); +int calculate_start_on_index( + TensorShape input_shape, int index, Coordinates starts, Coordinates strides, int32_t begin_mask); /** Returns the absolute end position of a given index for a strided slice operation * @@ -68,8 +69,13 @@ int calculate_start_on_index(TensorShape input_shape, int index, Coordinates sta * * @return Absolute end position of a given index */ -int calculate_end_on_index(TensorShape input_shape, int index, int start_on_index, Coordinates ends, Coordinates strides, - int32_t end_mask = 0, int32_t shrink_axis_mask = 0); +int calculate_end_on_index(TensorShape input_shape, + int index, + int start_on_index, + Coordinates ends, + Coordinates strides, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Calculate start, end and stride coordinates for a strided slice * @@ -87,8 +93,12 @@ int calculate_end_on_index(TensorShape input_shape, int index, int start_on_inde * @return A tuple with <Start,End,Strides> */ std::tuple<Coordinates, Coordinates, Coordinates> calculate_strided_slice_coords(TensorShape input_shape, - Coordinates starts, Coordinates ends, Coordinates strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + Coordinates starts, + Coordinates ends, + Coordinates strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Computes output shape of strided slice * @@ -109,9 +119,14 @@ std::tuple<Coordinates, Coordinates, Coordinates> calculate_strided_slice_coords * * @return The output tensor shape */ -TensorShape compute_strided_slice_output_shape(TensorShape input_shape, Coordinates starts, Coordinates ends, Coordinates strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0, - bool return_unshrinked = false); +TensorShape compute_strided_slice_output_shape(TensorShape input_shape, + Coordinates starts, + Coordinates ends, + Coordinates strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0, + bool return_unshrinked = false); /** Constructs end mask in case we want to perform a slice operation using the strided slice interface * @@ -122,7 +137,7 @@ TensorShape compute_strided_slice_output_shape(TensorShape input_shape, Coordina * @return End mask */ int32_t construct_slice_end_mask(Coordinates ends); -} // namespace tensor_tranform +} // namespace tensor_transform } // namespace helpers } // namespace arm_compute #endif /* ARM_COMPUTE_UTILS_HELPERS_TENSOR_TRANSFORM_H */ diff --git a/arm_compute/core/utils/logging/FilePrinter.h b/arm_compute/core/utils/logging/FilePrinter.h index 0e5b84f084..a865aadddb 100644 --- a/arm_compute/core/utils/logging/FilePrinter.h +++ b/arm_compute/core/utils/logging/FilePrinter.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_LOGGING_FILE_PRINTER_H #define ARM_COMPUTE_LOGGING_FILE_PRINTER_H -#include "arm_compute/core/utils/logging/IPrinter.h" - #include "arm_compute/core/utils/io/FileHandler.h" +#include "arm_compute/core/utils/logging/IPrinter.h" namespace arm_compute { diff --git a/arm_compute/core/utils/logging/Helpers.h b/arm_compute/core/utils/logging/Helpers.h index 5f8b948592..c3c2f0f0b8 100644 --- a/arm_compute/core/utils/logging/Helpers.h +++ b/arm_compute/core/utils/logging/Helpers.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_LOGGING_HELPERS_H #include "arm_compute/core/utils/logging/Types.h" + #include "support/ToolchainSupport.h" #include <cstddef> @@ -45,7 +46,7 @@ namespace logging * @return The formatted string */ template <typename... Ts> -inline std::string string_with_format(const std::string &fmt, Ts &&... args) +inline std::string string_with_format(const std::string &fmt, Ts &&...args) { size_t size = support::cpp11::snprintf(nullptr, 0, fmt.c_str(), args...) + 1; auto char_str = std::make_unique<char[]>(size); diff --git a/arm_compute/core/utils/logging/IPrinter.h b/arm_compute/core/utils/logging/IPrinter.h index 42dca58ea1..7fde4d9302 100644 --- a/arm_compute/core/utils/logging/IPrinter.h +++ b/arm_compute/core/utils/logging/IPrinter.h @@ -35,8 +35,7 @@ class Printer { public: /** Default Constructor */ - Printer() noexcept - : _mtx() + Printer() noexcept : _mtx() { } /** Prevent instances of this class from being copied */ diff --git a/arm_compute/core/utils/logging/LogMsgDecorators.h b/arm_compute/core/utils/logging/LogMsgDecorators.h index 9c9e62740f..66a8180e21 100644 --- a/arm_compute/core/utils/logging/LogMsgDecorators.h +++ b/arm_compute/core/utils/logging/LogMsgDecorators.h @@ -63,8 +63,7 @@ public: * * @param str Sting to append */ - StringDecorator(const std::string &str) - : _str(str) + StringDecorator(const std::string &str) : _str(str) { _str = angle_wrap_value(str); } @@ -103,7 +102,7 @@ private: auto time = std::chrono::system_clock::to_time_t(now); // TODO: use put_time for gcc > 4.9 - char buf[100] = { 0 }; + char buf[100] = {0}; std::strftime(buf, sizeof(buf), "%d-%m-%Y %I:%M:%S", std::localtime(&time)); return buf; } diff --git a/arm_compute/core/utils/logging/Logger.h b/arm_compute/core/utils/logging/Logger.h index 4fc9bb7dbf..608db39138 100644 --- a/arm_compute/core/utils/logging/Logger.h +++ b/arm_compute/core/utils/logging/Logger.h @@ -88,7 +88,7 @@ public: * @param[in] args Message arguments */ template <typename... Ts> - void log(LogLevel log_level, const std::string &fmt, Ts &&... args); + void log(LogLevel log_level, const std::string &fmt, Ts &&...args); /** Sets log level of the logger * * @warning Not thread-safe @@ -159,11 +159,11 @@ private: }; template <typename... Ts> -inline void Logger::log(LogLevel log_level, const std::string &fmt, Ts &&... args) +inline void Logger::log(LogLevel log_level, const std::string &fmt, Ts &&...args) { // Return if message shouldn't be logged // i.e. if log level does not match the logger's - if(!is_loggable(log_level)) + if (!is_loggable(log_level)) { return; } diff --git a/arm_compute/core/utils/logging/LoggerRegistry.h b/arm_compute/core/utils/logging/LoggerRegistry.h index 7c9931a260..4e52a10935 100644 --- a/arm_compute/core/utils/logging/LoggerRegistry.h +++ b/arm_compute/core/utils/logging/LoggerRegistry.h @@ -27,6 +27,7 @@ #include "arm_compute/core/utils/logging/Logger.h" #include "arm_compute/core/utils/logging/Printers.h" #include "arm_compute/core/utils/logging/Types.h" + #include "support/Mutex.h" #include <memory> @@ -54,8 +55,9 @@ public: * @param[in] log_level Logger's log level. Defaults to INFO * @param[in] printers Printers to attach to the system loggers. Defaults with a @ref StdPrinter. */ - void create_logger(const std::string &name, LogLevel log_level = LogLevel::INFO, - const std::vector<std::shared_ptr<Printer>> &printers = { std::make_shared<StdPrinter>() }); + void create_logger(const std::string &name, + LogLevel log_level = LogLevel::INFO, + const std::vector<std::shared_ptr<Printer>> &printers = {std::make_shared<StdPrinter>()}); /** Remove a logger * * @param name Logger's name @@ -74,16 +76,17 @@ public: * @param[in] printers (Optional) Printers to attach to the system loggers. Defaults with a @ref StdPrinter. */ void create_reserved_loggers(LogLevel log_level = LogLevel::INFO, - const std::vector<std::shared_ptr<Printer>> &printers = { std::make_shared<StdPrinter>() }); + const std::vector<std::shared_ptr<Printer>> &printers = { + std::make_shared<StdPrinter>()}); private: /** Default constructor */ LoggerRegistry(); private: - arm_compute::Mutex _mtx; + arm_compute::Mutex _mtx; std::unordered_map<std::string, std::shared_ptr<Logger>> _loggers; - static std::set<std::string> _reserved_loggers; + static std::set<std::string> _reserved_loggers; }; } // namespace logging } // namespace arm_compute diff --git a/arm_compute/core/utils/logging/Macros.h b/arm_compute/core/utils/logging/Macros.h index 0ab17c4464..4d5aa5fe2c 100644 --- a/arm_compute/core/utils/logging/Macros.h +++ b/arm_compute/core/utils/logging/Macros.h @@ -48,48 +48,48 @@ inline std::string signature_name(const std::string &pretty_func) do \ { \ auto __logger = arm_compute::logging::LoggerRegistry::get().logger(logger_name); \ - if(__logger != nullptr) \ + if (__logger != nullptr) \ { \ __logger->log(log_level, msg); \ } \ - } while(false) + } while (false) #define ARM_COMPUTE_LOG_MSG_WITH_FUNCNAME(logger_name, log_level, msg) \ do \ { \ auto __logger = arm_compute::logging::LoggerRegistry::get().logger(logger_name); \ - if(__logger != nullptr) \ + if (__logger != nullptr) \ { \ std::ostringstream s; \ s << ARM_COMPUTE_SIGNATURE_NAME << " : " << msg; \ __logger->log(log_level, s.str()); \ } \ - } while(false) + } while (false) #define ARM_COMPUTE_LOG_MSG_WITH_FORMAT(logger_name, log_level, fmt, ...) \ do \ { \ auto __logger = arm_compute::logging::LoggerRegistry::get().logger(logger_name); \ - if(__logger != nullptr) \ + if (__logger != nullptr) \ { \ size_t size = ::snprintf(nullptr, 0, fmt, __VA_ARGS__) + 1; \ auto char_str = std::make_unique<char[]>(size); \ ::snprintf(char_str.get(), size, fmt, __VA_ARGS__); \ __logger->log(log_level, std::string(char_str.get(), char_str.get() + size - 1)); \ } \ - } while(false) + } while (false) #define ARM_COMPUTE_LOG_STREAM(logger_name, log_level, stream) \ do \ { \ auto __logger = arm_compute::logging::LoggerRegistry::get().logger(logger_name); \ - if(__logger != nullptr) \ + if (__logger != nullptr) \ { \ std::ostringstream s; \ s << stream; \ __logger->log(log_level, s.str()); \ } \ - } while(false) + } while (false) #else /* ARM_COMPUTE_LOGGING_ENABLED */ diff --git a/arm_compute/core/utils/logging/Types.h b/arm_compute/core/utils/logging/Types.h index f0ddae6c84..64c567b984 100644 --- a/arm_compute/core/utils/logging/Types.h +++ b/arm_compute/core/utils/logging/Types.h @@ -44,8 +44,7 @@ enum class LogLevel struct LogMsg { /** Default constructor */ - LogMsg() - : raw_(), log_level_(LogLevel::OFF) + LogMsg() : raw_(), log_level_(LogLevel::OFF) { } /** Construct a log message @@ -53,8 +52,7 @@ struct LogMsg * @param[in] msg Message to log. * @param[in] log_level Logging level. Default: OFF */ - LogMsg(std::string msg, LogLevel log_level = LogLevel::OFF) - : raw_(msg), log_level_(log_level) + LogMsg(std::string msg, LogLevel log_level = LogLevel::OFF) : raw_(msg), log_level_(log_level) { } diff --git a/arm_compute/core/utils/math/Math.h b/arm_compute/core/utils/math/Math.h index c1dce7ff08..e70337ba0f 100644 --- a/arm_compute/core/utils/math/Math.h +++ b/arm_compute/core/utils/math/Math.h @@ -67,5 +67,5 @@ inline auto floor_to_multiple(S value, T divisor) -> decltype((value / divisor) return (value / divisor) * divisor; } -} +} // namespace arm_compute #endif /*ARM_COMPUTE_UTILS_MATH_H */ diff --git a/arm_compute/core/utils/math/SafeOps.h b/arm_compute/core/utils/math/SafeOps.h index dc928a0e5d..ef8bcf7e14 100644 --- a/arm_compute/core/utils/math/SafeOps.h +++ b/arm_compute/core/utils/math/SafeOps.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_UTILS_MATH_SAFE_OPS #include "arm_compute/core/Error.h" + #include "support/AclRequires.h" #include <limits> @@ -51,11 +52,11 @@ T safe_integer_add(T val_a, T val_b) { T result = 0; - if((val_b > 0) && (val_a > std::numeric_limits<T>::max() - val_b)) + if ((val_b > 0) && (val_a > std::numeric_limits<T>::max() - val_b)) { result = std::numeric_limits<T>::max(); } - else if((val_b < 0) && (val_a < std::numeric_limits<T>::min() - val_b)) + else if ((val_b < 0) && (val_a < std::numeric_limits<T>::min() - val_b)) { result = std::numeric_limits<T>::min(); } @@ -83,11 +84,11 @@ T safe_integer_sub(T val_a, T val_b) { T result = 0; - if((val_b < 0) && (val_a > std::numeric_limits<T>::max() + val_b)) + if ((val_b < 0) && (val_a > std::numeric_limits<T>::max() + val_b)) { result = std::numeric_limits<T>::max(); } - else if((val_b > 0) && (val_a < std::numeric_limits<T>::min() + val_b)) + else if ((val_b > 0) && (val_a < std::numeric_limits<T>::min() + val_b)) { result = std::numeric_limits<T>::min(); } @@ -115,13 +116,13 @@ T safe_integer_mul(T val_a, T val_b) { T result = 0; - if(val_a > 0) + if (val_a > 0) { - if((val_b > 0) && (val_a > (std::numeric_limits<T>::max() / val_b))) + if ((val_b > 0) && (val_a > (std::numeric_limits<T>::max() / val_b))) { result = std::numeric_limits<T>::max(); } - else if(val_b < (std::numeric_limits<T>::min() / val_a)) + else if (val_b < (std::numeric_limits<T>::min() / val_a)) { result = std::numeric_limits<T>::min(); } @@ -132,11 +133,11 @@ T safe_integer_mul(T val_a, T val_b) } else { - if((val_b > 0) && (val_a < (std::numeric_limits<T>::min() / val_b))) + if ((val_b > 0) && (val_a < (std::numeric_limits<T>::min() / val_b))) { result = std::numeric_limits<T>::max(); } - else if((val_a != 0) && (val_b < (std::numeric_limits<T>::max() / val_a))) + else if ((val_a != 0) && (val_b < (std::numeric_limits<T>::max() / val_a))) { result = std::numeric_limits<T>::min(); } @@ -165,7 +166,7 @@ T safe_integer_div(T val_a, T val_b) { T result = 0; - if((val_b == 0) || ((val_a == std::numeric_limits<T>::min()) && (val_b == -1))) + if ((val_b == 0) || ((val_a == std::numeric_limits<T>::min()) && (val_b == -1))) { result = std::numeric_limits<T>::min(); } @@ -176,7 +177,7 @@ T safe_integer_div(T val_a, T val_b) return result; } -} // namespace cast +} // namespace math } // namespace utils } // namespace arm_compute #endif /* ARM_COMPUTE_UTILS_MATH_SAFE_OPS */ diff --git a/arm_compute/core/utils/misc/InfoHelpers.h b/arm_compute/core/utils/misc/InfoHelpers.h index ced0d24b56..1d1b4ea8d7 100644 --- a/arm_compute/core/utils/misc/InfoHelpers.h +++ b/arm_compute/core/utils/misc/InfoHelpers.h @@ -53,10 +53,12 @@ inline bool is_relu(ActivationLayerInfo activation_info) */ inline bool is_relu6(ActivationLayerInfo activation_info) { - const bool is_lu_bounded_relu = activation_info.activation() == ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU - && activation_info.a() == 6.f && activation_info.b() == 0.f; - const bool is_bounded_relu = activation_info.activation() == ActivationLayerInfo::ActivationFunction::BOUNDED_RELU - && activation_info.a() == 6.f; + const bool is_lu_bounded_relu = + activation_info.activation() == ActivationLayerInfo::ActivationFunction::LU_BOUNDED_RELU && + activation_info.a() == 6.f && activation_info.b() == 0.f; + const bool is_bounded_relu = + activation_info.activation() == ActivationLayerInfo::ActivationFunction::BOUNDED_RELU && + activation_info.a() == 6.f; return activation_info.enabled() && (is_lu_bounded_relu || is_bounded_relu); } @@ -68,34 +70,37 @@ inline bool is_relu6(ActivationLayerInfo activation_info) * */ template <typename T> -inline void build_lstm_params_tensor_info(const LSTMParams<T> &lstm_params, - LSTMParams<ITensorInfo> *lstm_params_info) +inline void build_lstm_params_tensor_info(const LSTMParams<T> &lstm_params, LSTMParams<ITensorInfo> *lstm_params_info) { - if(lstm_params.has_peephole_opt()) + if (lstm_params.has_peephole_opt()) { ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.cell_to_forget_weights(), lstm_params.cell_to_output_weights()); - lstm_params_info->set_peephole_params(lstm_params.cell_to_forget_weights()->info(), lstm_params.cell_to_output_weights()->info()); + lstm_params_info->set_peephole_params(lstm_params.cell_to_forget_weights()->info(), + lstm_params.cell_to_output_weights()->info()); } - if(lstm_params.has_projection()) + if (lstm_params.has_projection()) { ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.projection_weights()); - lstm_params_info->set_projection_params(lstm_params.projection_weights()->info(), - lstm_params.projection_bias() != nullptr ? lstm_params.projection_bias()->info() : nullptr); + lstm_params_info->set_projection_params( + lstm_params.projection_weights()->info(), + lstm_params.projection_bias() != nullptr ? lstm_params.projection_bias()->info() : nullptr); } - if(!lstm_params.has_cifg_opt()) + if (!lstm_params.has_cifg_opt()) { - ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.input_to_input_weights(), lstm_params.recurrent_to_input_weights(), lstm_params.input_gate_bias()); + ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.input_to_input_weights(), lstm_params.recurrent_to_input_weights(), + lstm_params.input_gate_bias()); - ITensorInfo *cell_to_input_weights_info = (lstm_params.has_peephole_opt()) ? lstm_params.cell_to_input_weights()->info() : nullptr; - lstm_params_info->set_cifg_params(lstm_params.input_to_input_weights()->info(), lstm_params.recurrent_to_input_weights()->info(), - cell_to_input_weights_info, lstm_params.input_gate_bias()->info()); + ITensorInfo *cell_to_input_weights_info = + (lstm_params.has_peephole_opt()) ? lstm_params.cell_to_input_weights()->info() : nullptr; + lstm_params_info->set_cifg_params(lstm_params.input_to_input_weights()->info(), + lstm_params.recurrent_to_input_weights()->info(), cell_to_input_weights_info, + lstm_params.input_gate_bias()->info()); } - if(lstm_params.use_layer_norm()) + if (lstm_params.use_layer_norm()) { - ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.forget_layer_norm_weights(), - lstm_params.output_layer_norm_weights(), + ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.forget_layer_norm_weights(), lstm_params.output_layer_norm_weights(), lstm_params.cell_layer_norm_weights()); - if(!lstm_params.has_cifg_opt()) + if (!lstm_params.has_cifg_opt()) { ARM_COMPUTE_ERROR_ON_NULLPTR(lstm_params.input_layer_norm_weights()); } @@ -103,15 +108,14 @@ inline void build_lstm_params_tensor_info(const LSTMParams<T> &lstm_params, ITensorInfo *forget_info = lstm_params.forget_layer_norm_weights()->info(); ITensorInfo *cell_info = lstm_params.cell_layer_norm_weights()->info(); ITensorInfo *output_info = lstm_params.output_layer_norm_weights()->info(); - ITensorInfo *input_info = lstm_params.has_cifg_opt() ? nullptr : lstm_params.input_layer_norm_weights()->info(); + ITensorInfo *input_info = lstm_params.has_cifg_opt() ? nullptr : lstm_params.input_layer_norm_weights()->info(); lstm_params_info->set_layer_normalization_params(input_info, forget_info, cell_info, output_info); } - lstm_params_info->set_matmul_scale_params(lstm_params.input_intermediate_scale(), - lstm_params.forget_intermediate_scale(), - lstm_params.cell_intermediate_scale(), - lstm_params.output_intermediate_scale()); + lstm_params_info->set_matmul_scale_params( + lstm_params.input_intermediate_scale(), lstm_params.forget_intermediate_scale(), + lstm_params.cell_intermediate_scale(), lstm_params.output_intermediate_scale()); lstm_params_info->set_hidden_state_params(lstm_params.hidden_state_zero(), lstm_params.hidden_state_scale()); } diff --git a/arm_compute/core/utils/misc/Macros.h b/arm_compute/core/utils/misc/Macros.h index de66b6a52f..fa861fa442 100644 --- a/arm_compute/core/utils/misc/Macros.h +++ b/arm_compute/core/utils/misc/Macros.h @@ -26,15 +26,16 @@ #if defined(__cplusplus) && (__cplusplus >= 201402L) -#define ARM_COMPUTE_DEPRECATED [[deprecated]] -#define ARM_COMPUTE_DEPRECATED_REL(rel) [[deprecated("Deprecated in : " #rel)]] +#define ARM_COMPUTE_DEPRECATED [[deprecated]] +#define ARM_COMPUTE_DEPRECATED_REL(rel) [[deprecated("Deprecated in : " #rel)]] #define ARM_COMPUTE_DEPRECATED_REL_REPLACE(rel, replace) [[deprecated("Deprecated in : " #rel " - Use : " #replace)]] #elif defined(__GNUC__) || defined(__clang__) -#define ARM_COMPUTE_DEPRECATED __attribute__((deprecated)) +#define ARM_COMPUTE_DEPRECATED __attribute__((deprecated)) #define ARM_COMPUTE_DEPRECATED_REL(rel) __attribute__((deprecated("Deprecated in : " #rel))) -#define ARM_COMPUTE_DEPRECATED_REL_REPLACE(rel, replace) __attribute__((deprecated("Deprecated in : " #rel " - Use : " #replace))) +#define ARM_COMPUTE_DEPRECATED_REL_REPLACE(rel, replace) \ + __attribute__((deprecated("Deprecated in : " #rel " - Use : " #replace))) #else // defined(__cplusplus) && (__cplusplus >= 201402L) diff --git a/arm_compute/core/utils/misc/ShapeCalculator.h b/arm_compute/core/utils/misc/ShapeCalculator.h index 4c2037ab8d..31362f1ac4 100644 --- a/arm_compute/core/utils/misc/ShapeCalculator.h +++ b/arm_compute/core/utils/misc/ShapeCalculator.h @@ -28,11 +28,10 @@ #include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/KernelDescriptors.h" #include "arm_compute/core/Utils.h" +#include "arm_compute/core/utils/helpers/tensor_transform.h" #include "arm_compute/function_info/ConvolutionInfo.h" #include "arm_compute/runtime/FunctionDescriptors.h" -#include "arm_compute/core/utils/helpers/tensor_transform.h" - #include <cmath> namespace arm_compute @@ -57,12 +56,12 @@ inline TensorShape calculate_reduce_mean_shape(ITensorInfo *input, const Coordin convert_negative_axis(axis_local, input_dims); TensorShape out_shape = input->tensor_shape(); // Configure reshape layer if we want to drop the dimensions - if(!keep_dims) + if (!keep_dims) { // We have to sort the reduction axis vectors in order for remove_dimension // to work properly std::sort(axis_local.begin(), axis_local.begin() + reduction_ops); - for(int i = 0; i < reduction_ops; ++i) + for (int i = 0; i < reduction_ops; ++i) { out_shape.remove_dimension(axis_local[i] - i, false); } @@ -70,7 +69,7 @@ inline TensorShape calculate_reduce_mean_shape(ITensorInfo *input, const Coordin } else { - for(int i = 0; i < reduction_ops; ++i) + for (int i = 0; i < reduction_ops; ++i) { out_shape.set(axis_local[i], 1); } @@ -86,7 +85,10 @@ inline TensorShape calculate_reduce_mean_shape(ITensorInfo *input, const Coordin * * @return the calculated shape */ -inline TensorShape compute_vector_to_tensor_output_shape(const TensorShape &input, size_t conv_w, size_t conv_h, const DataLayout &data_layout) +inline TensorShape compute_vector_to_tensor_output_shape(const TensorShape &input, + size_t conv_w, + size_t conv_h, + const DataLayout &data_layout) { const size_t idx_w = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const size_t idx_h = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); @@ -128,10 +130,12 @@ inline TensorShape compute_reorg_output_shape(const ITensorInfo &input, int32_t const size_t idx_channel = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::CHANNEL); ARM_COMPUTE_ERROR_ON(stride <= 0); - ARM_COMPUTE_ERROR_ON_MSG((input.tensor_shape()[idx_width] % stride != 0), "The width of the input tensor must be a multiple of stride"); - ARM_COMPUTE_ERROR_ON_MSG((input.tensor_shape()[idx_height] % stride != 0), "The height of the input tensor must be a multiple of stride"); + ARM_COMPUTE_ERROR_ON_MSG((input.tensor_shape()[idx_width] % stride != 0), + "The width of the input tensor must be a multiple of stride"); + ARM_COMPUTE_ERROR_ON_MSG((input.tensor_shape()[idx_height] % stride != 0), + "The height of the input tensor must be a multiple of stride"); - TensorShape output_shape{ input.tensor_shape() }; + TensorShape output_shape{input.tensor_shape()}; output_shape.set(idx_width, output_shape[idx_width] / stride); output_shape.set(idx_height, output_shape[idx_height] / stride); @@ -148,7 +152,8 @@ inline TensorShape compute_reorg_output_shape(const ITensorInfo &input, int32_t * * @return the calculated shape of the reshaped weights */ -inline TensorShape compute_weights_reshaped_shape(const ITensorInfo &weights, bool has_bias = false, unsigned int num_groups = 1) +inline TensorShape +compute_weights_reshaped_shape(const ITensorInfo &weights, bool has_bias = false, unsigned int num_groups = 1) { // Number of groups greater than one are only supported for NCHW data layout, and the number of weights must be a multiple of it. ARM_COMPUTE_ERROR_ON(num_groups == 0); @@ -156,14 +161,14 @@ inline TensorShape compute_weights_reshaped_shape(const ITensorInfo &weights, bo ARM_COMPUTE_ERROR_ON((weights.dimension(3) % num_groups) != 0); // Calculate output shape - TensorShape weights_reshaped{ weights.tensor_shape() }; + TensorShape weights_reshaped{weights.tensor_shape()}; weights_reshaped.set(3, weights_reshaped[3] / num_groups); weights_reshaped.collapse(3); const size_t tmp_dim = weights_reshaped[0]; weights_reshaped.set(0, weights_reshaped[1]); weights_reshaped.set(1, tmp_dim + (has_bias ? 1 : 0)); - if(weights.num_dimensions() < 5) + if (weights.num_dimensions() < 5) { weights_reshaped.set(2, num_groups); } @@ -179,7 +184,9 @@ inline TensorShape compute_weights_reshaped_shape(const ITensorInfo &weights, bo * * @return the calculated shape */ -inline TensorShape compute_lhs_reshaped_shape(const ITensorInfo &a, const GEMMLHSMatrixInfo &lhs_info, bool reinterpret_input_as_3d = false) +inline TensorShape compute_lhs_reshaped_shape(const ITensorInfo &a, + const GEMMLHSMatrixInfo &lhs_info, + bool reinterpret_input_as_3d = false) { ARM_COMPUTE_ERROR_ON(lhs_info.m0 == 0); ARM_COMPUTE_ERROR_ON(lhs_info.k0 == 0); @@ -200,11 +207,11 @@ inline TensorShape compute_lhs_reshaped_shape(const ITensorInfo &a, const GEMMLH const unsigned int output_width = block_size * num_horiz_blocks * lhs_info.v0; const unsigned int output_height = std::ceil(num_vert_blocks / static_cast<float>(lhs_info.v0)); - TensorShape lhs_shape{ a.tensor_shape() }; + TensorShape lhs_shape{a.tensor_shape()}; lhs_shape.set(0, output_width); lhs_shape.set(1, output_height); - if((reinterpret_input_as_3d) && (lhs_shape.num_dimensions() > 2)) + if ((reinterpret_input_as_3d) && (lhs_shape.num_dimensions() > 2)) { // When the data format is NHWC and the shapes are Nx1x1 // the tensor shape num_dimensions is automatically set to 1 instead of 3. @@ -244,7 +251,7 @@ inline TensorShape compute_rhs_reshaped_shape(const ITensorInfo &a, const GEMMRH const unsigned int output_width = block_size * num_vert_blocks * rhs_info.h0; const unsigned int output_height = std::ceil(num_horiz_blocks / static_cast<float>(rhs_info.h0)); - TensorShape rhs_shape{ a.tensor_shape() }; + TensorShape rhs_shape{a.tensor_shape()}; rhs_shape.set(0, output_width); rhs_shape.set(1, output_height); @@ -259,14 +266,15 @@ inline TensorShape compute_rhs_reshaped_shape(const ITensorInfo &a, const GEMMRH * * @return the calculated shape */ -inline TensorShape compute_interleaved_shape(const ITensorInfo &a, int mult_interleave4x4_height = 1, bool reinterpret_input_as_3d = false) +inline TensorShape +compute_interleaved_shape(const ITensorInfo &a, int mult_interleave4x4_height = 1, bool reinterpret_input_as_3d = false) { // The interleaved output matrix will have the following shape: [ a_height * W, ceil(a_width / W) ] where W = 4 * mult_interleave4x4_height ARM_COMPUTE_ERROR_ON(mult_interleave4x4_height < 1); const int interleave_width = 4 * mult_interleave4x4_height; - TensorShape shape_interleaved_a{ a.tensor_shape() }; + TensorShape shape_interleaved_a{a.tensor_shape()}; shape_interleaved_a.set(0, a.dimension(0) * interleave_width); - if(reinterpret_input_as_3d) + if (reinterpret_input_as_3d) { const int M = a.dimension(1) * a.dimension(2); const int height = std::ceil(M / static_cast<float>(interleave_width)); @@ -276,7 +284,7 @@ inline TensorShape compute_interleaved_shape(const ITensorInfo &a, int mult_inte // the tensor shape num_dimensions is automatically set to 1 instead of 3. // To avoid failures by removing a dimension that doesn't exist // check if the number of dimensions is greater than 2. - if(shape_interleaved_a.num_dimensions() > 2) + if (shape_interleaved_a.num_dimensions() > 2) { shape_interleaved_a.remove_dimension(2); } @@ -298,7 +306,7 @@ inline TensorShape compute_interleaved_shape(const ITensorInfo &a, int mult_inte inline TensorShape compute_transpose1xW_shape(const ITensorInfo &b) { // The transpose1xW output matrix will have the following shape: [ b_height * 16, ceil(b_width / 16.0f) ] - TensorShape shape_transposed1xW_b{ b.tensor_shape() }; + TensorShape shape_transposed1xW_b{b.tensor_shape()}; shape_transposed1xW_b.set(0, b.dimension(1) * 16); shape_transposed1xW_b.set(1, std::ceil(b.dimension(0) / 16.f)); @@ -318,7 +326,7 @@ inline TensorShape compute_transpose1xW_with_element_size_shape(const ITensorInf // The transpose1xW output matrix will have the following shape: // [ b_height * W, ceil(b_width / W) ] where W = (16 / element size of the tensor) * mult_transpose1xW_width ARM_COMPUTE_ERROR_ON(mult_transpose1xW_width < 1); - TensorShape shape_transposed1xW_b{ b.tensor_shape() }; + TensorShape shape_transposed1xW_b{b.tensor_shape()}; const size_t transpose_width = (16 / b.element_size()) * mult_transpose1xW_width; shape_transposed1xW_b.set(0, b.dimension(1) * transpose_width); shape_transposed1xW_b.set(1, static_cast<size_t>(std::ceil(b.dimension(0) / static_cast<float>(transpose_width)))); @@ -334,8 +342,8 @@ inline TensorShape compute_transpose1xW_with_element_size_shape(const ITensorInf */ inline TensorShape compute_reductionA_shape(const ITensorInfo &b) { - TensorShape shape_vector_sum_col{ b.tensor_shape() }; - if(shape_vector_sum_col.num_dimensions() > 1) + TensorShape shape_vector_sum_col{b.tensor_shape()}; + if (shape_vector_sum_col.num_dimensions() > 1) { shape_vector_sum_col.remove_dimension(1); } @@ -351,9 +359,9 @@ inline TensorShape compute_reductionA_shape(const ITensorInfo &b) */ inline TensorShape compute_reductionB_shape(const ITensorInfo &a) { - TensorShape shape_vector_sum_row{ a.tensor_shape() }; + TensorShape shape_vector_sum_row{a.tensor_shape()}; shape_vector_sum_row.set(Window::DimX, a.dimension(1)); - if(shape_vector_sum_row.num_dimensions() > 1) + if (shape_vector_sum_row.num_dimensions() > 1) { shape_vector_sum_row.remove_dimension(1); } @@ -370,7 +378,10 @@ inline TensorShape compute_reductionB_shape(const ITensorInfo &a) * * @return the calculated shape */ -inline TensorShape compute_col2im_shape(const ITensorInfo &input, const Size2D &convolved_dims, bool batch_size_on_z, unsigned int num_groups = 1) +inline TensorShape compute_col2im_shape(const ITensorInfo &input, + const Size2D &convolved_dims, + bool batch_size_on_z, + unsigned int num_groups = 1) { ARM_COMPUTE_ERROR_ON(num_groups == 0); ARM_COMPUTE_ERROR_ON(input.tensor_shape()[1] != (convolved_dims.area())); @@ -381,10 +392,10 @@ inline TensorShape compute_col2im_shape(const ITensorInfo &input, const Size2D & const int height_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const int channel_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); - TensorShape col2im_shape{ input.tensor_shape() }; + TensorShape col2im_shape{input.tensor_shape()}; // If batches start on 3rd dimension shift dimensions right by 1 to retain upper tensor shape, // as first three will be override by H,W,C data - if(batch_size_on_z && num_groups == 1) + if (batch_size_on_z && num_groups == 1) { col2im_shape.shift_right(1); } @@ -403,7 +414,7 @@ inline TensorShape compute_col2im_shape(const ITensorInfo &input, const Size2D & */ inline TensorShape compute_transposed_shape(const ITensorInfo &input) { - TensorShape shape_transposed{ input.tensor_shape() }; + TensorShape shape_transposed{input.tensor_shape()}; shape_transposed.set(0, input.dimension(1), false); shape_transposed.set(1, input.dimension(0), false); @@ -419,10 +430,11 @@ inline TensorShape compute_transposed_shape(const ITensorInfo &input) * * @return the calculated shape */ -inline TensorShape compute_depthwise_convolution_shape(const ITensorInfo &input, const ITensorInfo &weights, const ConvolutionInfo &info) +inline TensorShape +compute_depthwise_convolution_shape(const ITensorInfo &input, const ITensorInfo &weights, const ConvolutionInfo &info) { - const TensorShape input_shape{ input.tensor_shape() }; - const TensorShape weights_shape{ weights.tensor_shape() }; + const TensorShape input_shape{input.tensor_shape()}; + const TensorShape weights_shape{weights.tensor_shape()}; const DataLayout data_layout = input.data_layout(); const int width_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -430,16 +442,16 @@ inline TensorShape compute_depthwise_convolution_shape(const ITensorInfo &input, const int channel_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); const DataLayout weights_data_layout = weights.data_layout(); - const int weights_width_idx = get_data_layout_dimension_index(weights_data_layout, DataLayoutDimension::WIDTH); - const int weights_height_idx = get_data_layout_dimension_index(weights_data_layout, DataLayoutDimension::HEIGHT); + const int weights_width_idx = get_data_layout_dimension_index(weights_data_layout, DataLayoutDimension::WIDTH); + const int weights_height_idx = get_data_layout_dimension_index(weights_data_layout, DataLayoutDimension::HEIGHT); unsigned int output_width = 0; unsigned int output_height = 0; - std::tie(output_width, output_height) = scaled_dimensions(input_shape[width_idx], input_shape[height_idx], - weights_shape[weights_width_idx], weights_shape[weights_height_idx], - info.pad_stride_info, info.dilation); + std::tie(output_width, output_height) = + scaled_dimensions(input_shape[width_idx], input_shape[height_idx], weights_shape[weights_width_idx], + weights_shape[weights_height_idx], info.pad_stride_info, info.dilation); - TensorShape output_shape{ input_shape }; + TensorShape output_shape{input_shape}; output_shape.set(width_idx, output_width); output_shape.set(height_idx, output_height); output_shape.set(channel_idx, input_shape[channel_idx] * info.depth_multiplier); @@ -459,8 +471,13 @@ inline TensorShape compute_depthwise_convolution_shape(const ITensorInfo &input, * * @return the calculated shape */ -inline TensorShape compute_deconvolution_upsampled_shape(const ITensorInfo &input, const ITensorInfo &weights, unsigned int sx, unsigned int sy, - std::pair<unsigned int, unsigned int> &out_dims, uint32_t &padx, uint32_t &pady) +inline TensorShape compute_deconvolution_upsampled_shape(const ITensorInfo &input, + const ITensorInfo &weights, + unsigned int sx, + unsigned int sy, + std::pair<unsigned int, unsigned int> &out_dims, + uint32_t &padx, + uint32_t &pady) { const DataLayout data_layout = input.data_layout(); const size_t idx_w = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -491,10 +508,12 @@ inline TensorShape compute_deconvolution_upsampled_shape(const ITensorInfo &inpu * * @return the calculated shape */ -inline TensorShape compute_deconvolution_output_shape(const std::pair<unsigned int, unsigned int> &out_dims, const ITensorInfo &input, const ITensorInfo &weights) +inline TensorShape compute_deconvolution_output_shape(const std::pair<unsigned int, unsigned int> &out_dims, + const ITensorInfo &input, + const ITensorInfo &weights) { - const TensorShape input_shape{ input.tensor_shape() }; - const TensorShape weights_shape{ weights.tensor_shape() }; + const TensorShape input_shape{input.tensor_shape()}; + const TensorShape weights_shape{weights.tensor_shape()}; const DataLayout data_layout = input.data_layout(); const int width_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -502,7 +521,7 @@ inline TensorShape compute_deconvolution_output_shape(const std::pair<unsigned i const int channel_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); const int batch_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::BATCHES); - TensorShape out_shape{ input_shape }; + TensorShape out_shape{input_shape}; out_shape.set(width_idx, out_dims.first); out_shape.set(height_idx, out_dims.second); out_shape.set(channel_idx, weights_shape[batch_idx]); @@ -522,8 +541,14 @@ inline TensorShape compute_deconvolution_output_shape(const std::pair<unsigned i * * @return the calculated shape */ -inline TensorShape compute_im2col_conv_shape(const ITensorInfo *input, const Size2D &kernel_dims, const PadStrideInfo &conv_info, bool has_bias, const Size2D &dilation, bool batch_size_on_z, - unsigned int num_groups = 1, unsigned int input_pad_right = 0) +inline TensorShape compute_im2col_conv_shape(const ITensorInfo *input, + const Size2D &kernel_dims, + const PadStrideInfo &conv_info, + bool has_bias, + const Size2D &dilation, + bool batch_size_on_z, + unsigned int num_groups = 1, + unsigned int input_pad_right = 0) { // The output shape will be the 3D shape [ out_channels * kernel_area, num_elems_per_out_channel, batches ] if batch_size_on_z == true // or the 4D shape [ out_channels * kernel_area / num_groups, num_elems_per_out_channel, num_groups, batches ] if batch_size_on_z == false @@ -532,17 +557,19 @@ inline TensorShape compute_im2col_conv_shape(const ITensorInfo *input, const Siz ARM_COMPUTE_ERROR_ON(num_groups > 1 && input->data_layout() != DataLayout::NCHW); ARM_COMPUTE_ERROR_ON(num_groups > 1 && batch_size_on_z); - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; const DataLayout data_layout = input->data_layout(); const int width_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const int height_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const int channel_idx = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); - std::pair<unsigned int, unsigned int> out_dims = scaled_dimensions(output_shape[width_idx], output_shape[height_idx], kernel_dims.width, kernel_dims.height, conv_info, dilation); - output_shape.set(0, ((output_shape[channel_idx] + input_pad_right) / num_groups * kernel_dims.area() + (has_bias ? 1 : 0))); // NOLINT + std::pair<unsigned int, unsigned int> out_dims = scaled_dimensions( + output_shape[width_idx], output_shape[height_idx], kernel_dims.width, kernel_dims.height, conv_info, dilation); + output_shape.set(0, ((output_shape[channel_idx] + input_pad_right) / num_groups * kernel_dims.area() + + (has_bias ? 1 : 0))); // NOLINT output_shape.set(1, (out_dims.first * out_dims.second)); - if(batch_size_on_z && output_shape.num_dimensions() >= 3) + if (batch_size_on_z && output_shape.num_dimensions() >= 3) { output_shape.remove_dimension(2); } @@ -564,7 +591,7 @@ inline TensorShape compute_flatten_shape(const ITensorInfo *input) { // The output shape will be the flatten version of the input (i.e. [ width * height * channels, num_batches, ... ] ). Used for FlattenLayer and FullyConnectedLayer. - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; output_shape.collapse(3); @@ -586,7 +613,7 @@ inline TensorShape compute_softmax_shape(const ITensorInfo *input, size_t axis = // - [x,y,z,w] and axis 3 will return [x*y*z, w] TensorShape shape2D = input->tensor_shape(); - if(axis < input->num_dimensions()) + if (axis < input->num_dimensions()) { // Collapse from axis onward (this changes the shape) shape2D.collapse_from(axis); @@ -600,7 +627,7 @@ inline TensorShape compute_softmax_shape(const ITensorInfo *input, size_t axis = shape2D.collapse(shape2D.num_dimensions()); } - if(axis == 0) + if (axis == 0) { // If axis is zero the first dim should be one. Since // collapse is an inclusive operation we need to shift @@ -619,15 +646,17 @@ inline TensorShape compute_softmax_shape(const ITensorInfo *input, size_t axis = */ inline TensorShape compute_winograd_filter_transform_shape(const ITensorInfo &input, const WinogradInfo &winograd_info) { - TensorShape tensor_shape{ input.tensor_shape() }; + TensorShape tensor_shape{input.tensor_shape()}; const Size2D kernel_size = winograd_info.kernel_size; const Size2D output_tile_size = winograd_info.output_tile_size; - const Size2D input_tile_size = Size2D(output_tile_size.width + kernel_size.width - 1, output_tile_size.height + kernel_size.height - 1); + const Size2D input_tile_size = + Size2D(output_tile_size.width + kernel_size.width - 1, output_tile_size.height + kernel_size.height - 1); tensor_shape.remove_dimension(get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::WIDTH)); tensor_shape.set(Window::DimX, input.dimension(3)); - tensor_shape.set(Window::DimY, input.dimension(get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::CHANNEL))); + tensor_shape.set(Window::DimY, input.dimension(get_data_layout_dimension_index(input.data_layout(), + DataLayoutDimension::CHANNEL))); tensor_shape.set(Window::DimZ, input_tile_size.area()); return tensor_shape; @@ -645,23 +674,22 @@ inline TensorShape compute_winograd_input_transform_shape(const ITensorInfo &inp const PadStrideInfo conv_info = winograd_info.convolution_info; const Size2D kernel_size = winograd_info.kernel_size; const Size2D output_tile_size = winograd_info.output_tile_size; - const Size2D input_tile_size = Size2D(output_tile_size.width + kernel_size.width - 1, output_tile_size.height + kernel_size.height - 1); + const Size2D input_tile_size = + Size2D(output_tile_size.width + kernel_size.width - 1, output_tile_size.height + kernel_size.height - 1); const size_t idx_w = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::WIDTH); const size_t idx_h = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::HEIGHT); const size_t idx_c = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::CHANNEL); // Compute the number of output tiles along the x and y direction of size "output_tile_size" - const Size2D num_tiles = compute_winograd_convolution_tiles(Size2D(input.tensor_shape()[idx_w], input.tensor_shape()[idx_h]), - kernel_size, - output_tile_size, - conv_info); + const Size2D num_tiles = compute_winograd_convolution_tiles( + Size2D(input.tensor_shape()[idx_w], input.tensor_shape()[idx_h]), kernel_size, output_tile_size, conv_info); const unsigned int width = input.tensor_shape()[idx_c]; const unsigned int height = num_tiles.area(); const unsigned int depth = input_tile_size.area(); - TensorShape output_shape{ input.tensor_shape() }; + TensorShape output_shape{input.tensor_shape()}; output_shape.set(0, width); output_shape.set(1, height); output_shape.set(2, depth); @@ -684,12 +712,12 @@ inline TensorShape compute_winograd_output_transform_shape(const ITensorInfo &in const DataLayout data_layout = winograd_info.output_data_layout; // Compute output shape - unsigned int output_width = 0; - unsigned int output_height = 0; + unsigned int output_width = 0; + unsigned int output_height = 0; std::tie(output_width, output_height) = scaled_dimensions(input_dimensions.width, input_dimensions.height, kernel_size.width, kernel_size.height, conv_info); - TensorShape tensor_shape{ input.tensor_shape() }; + TensorShape tensor_shape{input.tensor_shape()}; // Output dimension const unsigned int out_w = output_width; @@ -712,7 +740,10 @@ inline TensorShape compute_winograd_output_transform_shape(const ITensorInfo &in * * @return the calculated shape */ -inline TensorShape compute_deep_convolution_shape(const TensorShape &input_shape, DataLayout input_data_layout, const TensorShape &weights_shape, const PadStrideInfo &conv_info) +inline TensorShape compute_deep_convolution_shape(const TensorShape &input_shape, + DataLayout input_data_layout, + const TensorShape &weights_shape, + const PadStrideInfo &conv_info) { const size_t idx_width = get_data_layout_dimension_index(input_data_layout, DataLayoutDimension::WIDTH); const size_t idx_height = get_data_layout_dimension_index(input_data_layout, DataLayoutDimension::HEIGHT); @@ -725,9 +756,10 @@ inline TensorShape compute_deep_convolution_shape(const TensorShape &input_shape const unsigned int weights_out_channel = weights_shape[3]; unsigned int output_width = 0; unsigned int output_height = 0; - std::tie(output_width, output_height) = scaled_dimensions(input_width, input_height, weights_width, weights_height, conv_info); + std::tie(output_width, output_height) = + scaled_dimensions(input_width, input_height, weights_width, weights_height, conv_info); - TensorShape output_shape{ input_shape }; + TensorShape output_shape{input_shape}; output_shape.set(idx_width, output_width); output_shape.set(idx_height, output_height); output_shape.set(idx_channel, weights_out_channel); @@ -743,7 +775,8 @@ inline TensorShape compute_deep_convolution_shape(const TensorShape &input_shape * * @return the calculated shape */ -inline TensorShape compute_deep_convolution_shape(const ITensorInfo &input, const ITensorInfo &weights, const PadStrideInfo &conv_info) +inline TensorShape +compute_deep_convolution_shape(const ITensorInfo &input, const ITensorInfo &weights, const PadStrideInfo &conv_info) { return compute_deep_convolution_shape(input.tensor_shape(), input.data_layout(), weights.tensor_shape(), conv_info); } @@ -758,7 +791,10 @@ inline TensorShape compute_deep_convolution_shape(const ITensorInfo &input, cons * * @return the calculated shape */ -inline TensorShape compute_indirect_buffer_shape(const TensorShape &input_shape, DataLayout input_data_layout, const TensorShape &weights_shape, const PadStrideInfo &conv_info, +inline TensorShape compute_indirect_buffer_shape(const TensorShape &input_shape, + DataLayout input_data_layout, + const TensorShape &weights_shape, + const PadStrideInfo &conv_info, const DirectConvComputeKernelInfo &desc) { ARM_COMPUTE_ERROR_ON_MSG(input_data_layout != DataLayout::NHWC, "The data layout can only be NHWC"); @@ -768,7 +804,8 @@ inline TensorShape compute_indirect_buffer_shape(const TensorShape &input_shape, const unsigned int kw = weights_shape[1]; const unsigned int kh = weights_shape[2]; - TensorShape output_conv2d_shape = compute_deep_convolution_shape(input_shape, input_data_layout, weights_shape, conv_info); + TensorShape output_conv2d_shape = + compute_deep_convolution_shape(input_shape, input_data_layout, weights_shape, conv_info); const unsigned int output_w = m0 * kw * kh; const unsigned int output_h = DIV_CEIL(output_conv2d_shape[1] * output_conv2d_shape[2], m0); @@ -785,7 +822,7 @@ inline TensorShape compute_indirect_buffer_shape(const TensorShape &input_shape, */ inline TensorShape compute_min_max_shape(const ITensorInfo *input) { - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; output_shape.set(Window::DimX, 2); output_shape.remove_dimension(1); output_shape.remove_dimension(1); @@ -805,7 +842,7 @@ inline TensorShape compute_pool_shape(const ITensorInfo &input, PoolingLayerInfo int pooled_w = 0; int pooled_h = 0; - TensorShape output_shape{ input.tensor_shape() }; + TensorShape output_shape{input.tensor_shape()}; const bool is_global_pooling = pool_info.is_global_pooling; const int idx_width = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::WIDTH); @@ -815,9 +852,8 @@ inline TensorShape compute_pool_shape(const ITensorInfo &input, PoolingLayerInfo const int pool_size_x = is_global_pooling ? output_shape[idx_width] : pool_info.pool_size.width; const int pool_size_y = is_global_pooling ? output_shape[idx_height] : pool_info.pool_size.height; - std::tie(pooled_w, pooled_h) = scaled_dimensions_signed(input_width, input_height, - pool_size_x, pool_size_y, - pool_info.pad_stride_info); + std::tie(pooled_w, pooled_h) = + scaled_dimensions_signed(input_width, input_height, pool_size_x, pool_size_y, pool_info.pad_stride_info); ARM_COMPUTE_ERROR_ON_MSG((pooled_w < 1 || pooled_h < 1), "Calculated output dimension size is invalid"); @@ -850,8 +886,10 @@ inline TensorShape compute_unpool_shape(const ITensorInfo &input, PoolingLayerIn const int pad_bottom = pad_stride_info.pad_bottom(); TensorShape output_shape = input_shape; - const unsigned int out_width = (input_shape[idx_width] - 1) * stride_x - pad_left - pad_right + pool_info.pool_size.width; - const unsigned int out_height = (input_shape[idx_height] - 1) * stride_y - pad_top - pad_bottom + pool_info.pool_size.height; + const unsigned int out_width = + (input_shape[idx_width] - 1) * stride_x - pad_left - pad_right + pool_info.pool_size.width; + const unsigned int out_height = + (input_shape[idx_height] - 1) * stride_y - pad_top - pad_bottom + pool_info.pool_size.height; output_shape.set(idx_width, out_width); output_shape.set(idx_height, out_height); @@ -866,9 +904,10 @@ inline TensorShape compute_unpool_shape(const ITensorInfo &input, PoolingLayerIn * * @return the calculated shape */ -inline TensorShape compute_roi_align_shape(const ITensorInfo &input, const ITensorInfo &rois, ROIPoolingLayerInfo pool_info) +inline TensorShape +compute_roi_align_shape(const ITensorInfo &input, const ITensorInfo &rois, ROIPoolingLayerInfo pool_info) { - TensorShape output_shape{ input.tensor_shape() }; + TensorShape output_shape{input.tensor_shape()}; const unsigned int idx_width = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::WIDTH); const unsigned int idx_height = get_data_layout_dimension_index(input.data_layout(), DataLayoutDimension::HEIGHT); @@ -889,7 +928,7 @@ inline TensorShape compute_roi_align_shape(const ITensorInfo &input, const ITens */ inline TensorShape compute_rnn_shape(const ITensorInfo *input, const unsigned int batch_size) { - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; output_shape.set(1, batch_size); return output_shape; @@ -904,15 +943,21 @@ inline TensorShape compute_rnn_shape(const ITensorInfo *input, const unsigned in * * @return the calculated shape */ -inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, bool is_interleaved_transposed, const GEMMReshapeInfo &reshape_info) +inline TensorShape compute_mm_shape(const ITensorInfo &input0, + const ITensorInfo &input1, + bool is_interleaved_transposed, + const GEMMReshapeInfo &reshape_info) { ARM_COMPUTE_ERROR_ON_MSG(input0.num_dimensions() > 4, "The number of dimensions for the matrix A must be <= 4"); - ARM_COMPUTE_ERROR_ON_MSG(is_interleaved_transposed && reshape_info.reinterpret_input_as_3d(), "The first input tensor cannot be reinterpreted as 3D if is_interleaved_transposed is true"); + ARM_COMPUTE_ERROR_ON_MSG( + is_interleaved_transposed && reshape_info.reinterpret_input_as_3d(), + "The first input tensor cannot be reinterpreted as 3D if is_interleaved_transposed is true"); const bool reinterpret_input_as_3d = reshape_info.reinterpret_input_as_3d(); const bool reinterpret_output_as_3d = reshape_info.depth_output_gemm3d() != 0; const int depth_output_gemm3d = reinterpret_output_as_3d ? reshape_info.depth_output_gemm3d() : 1; - const int m = reshape_info.reinterpret_input_as_3d() ? input0.dimension(1) * input0.dimension(2) : input0.dimension(1); + const int m = + reshape_info.reinterpret_input_as_3d() ? input0.dimension(1) * input0.dimension(2) : input0.dimension(1); // If the output of GEMM has to be reinterpreted as 3D, the number of input0 rows (M) is obtained collapsing the second and third // dimension of the output tensor @@ -921,7 +966,7 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo const int dim2 = reinterpret_input_as_3d ? input0.tensor_shape()[3] : input0.tensor_shape()[2]; const int dim3 = reinterpret_input_as_3d ? 1 : input0.tensor_shape()[3]; - TensorShape output_shape{ input0.tensor_shape() }; + TensorShape output_shape{input0.tensor_shape()}; output_shape.set(0, dim0); output_shape.set(1, dim1); @@ -940,7 +985,8 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo * * @return the calculated shape */ -inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, const GEMMReshapeInfo &gemm_info) +inline TensorShape +compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, const GEMMReshapeInfo &gemm_info) { ARM_COMPUTE_UNUSED(input1); ARM_COMPUTE_ERROR_ON_MSG(input0.num_dimensions() > 4, "The number of dimensions for the matrix A must be <= 4"); @@ -949,9 +995,9 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo const bool reinterpret_output_as_3d = gemm_info.depth_output_gemm3d() != 0; const int depth_output_gemm3d = reinterpret_output_as_3d ? gemm_info.depth_output_gemm3d() : 1; - TensorShape output_shape{ input0.tensor_shape() }; + TensorShape output_shape{input0.tensor_shape()}; - if(!reinterpret_input_as_3d && !reinterpret_output_as_3d) + if (!reinterpret_input_as_3d && !reinterpret_output_as_3d) { output_shape.set(0, gemm_info.n()); output_shape.set(1, gemm_info.m()); @@ -978,7 +1024,8 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo * * @return the calculated shape */ -inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, const GEMMKernelInfo &gemm_info) +inline TensorShape +compute_mm_shape(const ITensorInfo &input0, const ITensorInfo &input1, const GEMMKernelInfo &gemm_info) { ARM_COMPUTE_UNUSED(input1); ARM_COMPUTE_ERROR_ON_MSG(input0.num_dimensions() > 4, "The number of dimensions for the matrix A must be <= 4"); @@ -987,9 +1034,9 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo const bool reinterpret_output_as_3d = gemm_info.depth_output_gemm3d != 0; const unsigned int depth_output_gemm3d = reinterpret_output_as_3d ? gemm_info.depth_output_gemm3d : 1; - TensorShape output_shape{ input0.tensor_shape() }; + TensorShape output_shape{input0.tensor_shape()}; - if(!reinterpret_input_as_3d && !reinterpret_output_as_3d) + if (!reinterpret_input_as_3d && !reinterpret_output_as_3d) { output_shape.set(0, gemm_info.n); output_shape.set(1, gemm_info.m); @@ -1016,16 +1063,17 @@ inline TensorShape compute_mm_shape(const ITensorInfo &input0, const ITensorInfo * * @return the calculated shape */ -inline TensorShape compute_matmul_shape(const TensorShape &input0, const TensorShape &input1, const MatMulKernelInfo &matmul_info) +inline TensorShape +compute_matmul_shape(const TensorShape &input0, const TensorShape &input1, const MatMulKernelInfo &matmul_info) { - TensorShape output_shape{ input0 }; + TensorShape output_shape{input0}; - if(matmul_info.adj_lhs) + if (matmul_info.adj_lhs) { output_shape.set(1, input0[0]); // The vertical (M) dimension } - if(matmul_info.adj_rhs) + if (matmul_info.adj_rhs) { output_shape.set(0, input1[1]); // The horizontal (N) dimension } @@ -1044,14 +1092,15 @@ inline TensorShape compute_matmul_shape(const TensorShape &input0, const TensorS * * @return the calculated shape */ -inline TensorShape compute_output_stage_shape(const ITensorInfo &input, unsigned int gemm_3d_depth = 1, bool batch_size_on_z = false) +inline TensorShape +compute_output_stage_shape(const ITensorInfo &input, unsigned int gemm_3d_depth = 1, bool batch_size_on_z = false) { ARM_COMPUTE_ERROR_ON(input.data_layout() != DataLayout::NHWC && gemm_3d_depth > 1); TensorShape output_shape = input.tensor_shape(); - if(gemm_3d_depth > 1) + if (gemm_3d_depth > 1) { - if(batch_size_on_z) + if (batch_size_on_z) { output_shape.shift_right(1); } @@ -1076,11 +1125,16 @@ inline TensorShape compute_output_stage_shape(const ITensorInfo &input, unsigned * @return the calculated shape */ inline TensorShape compute_strided_slice_shape(const ITensorInfo &input, - const Coordinates &starts, const Coordinates &ends, const Coordinates &strides, - int32_t begin_mask, int32_t end_mask, int32_t shrink_axis_mask) + const Coordinates &starts, + const Coordinates &ends, + const Coordinates &strides, + int32_t begin_mask, + int32_t end_mask, + int32_t shrink_axis_mask) { using namespace arm_compute::helpers::tensor_transform; - return compute_strided_slice_output_shape(input.tensor_shape(), starts, ends, strides, begin_mask, end_mask, shrink_axis_mask); + return compute_strided_slice_output_shape(input.tensor_shape(), starts, ends, strides, begin_mask, end_mask, + shrink_axis_mask); } /** Calculate the slice output shape of a tensor @@ -1091,13 +1145,13 @@ inline TensorShape compute_strided_slice_shape(const ITensorInfo &input, * * @return the calculated shape */ -inline TensorShape compute_slice_shape(const TensorShape &input_shape, const Coordinates &starts, const Coordinates &ends) +inline TensorShape +compute_slice_shape(const TensorShape &input_shape, const Coordinates &starts, const Coordinates &ends) { using namespace arm_compute::helpers::tensor_transform; - return compute_strided_slice_output_shape(input_shape, - starts, ends, BiStrides(), - 0, construct_slice_end_mask(ends), 0); + return compute_strided_slice_output_shape(input_shape, starts, ends, BiStrides(), 0, construct_slice_end_mask(ends), + 0); } /** Calculate the batch to space output shape of a tensor @@ -1110,7 +1164,8 @@ inline TensorShape compute_slice_shape(const TensorShape &input_shape, const Coo * * @return the calculated shape */ -inline TensorShape compute_batch_to_space_shape(DataLayout data_layout, const TensorShape &input, int block_x, int block_y, const CropInfo &crop_info = CropInfo{}) +inline TensorShape compute_batch_to_space_shape( + DataLayout data_layout, const TensorShape &input, int block_x, int block_y, const CropInfo &crop_info = CropInfo{}) { ARM_COMPUTE_ERROR_ON(block_x < 1 || block_y < 1); @@ -1118,7 +1173,7 @@ inline TensorShape compute_batch_to_space_shape(DataLayout data_layout, const Te const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const int idx_batch = get_data_layout_dimension_index(data_layout, DataLayoutDimension::BATCHES); - TensorShape output_shape{ input }; + TensorShape output_shape{input}; unsigned int new_width = input[idx_width] * static_cast<unsigned int>(block_x); unsigned int new_height = input[idx_height] * static_cast<unsigned int>(block_y); @@ -1152,7 +1207,7 @@ inline TensorShape compute_depth_to_space_shape(const TensorShape &input_shape, const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const int idx_channel = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); - TensorShape output_shape{ input_shape }; + TensorShape output_shape{input_shape}; output_shape.set(idx_width, input_shape[idx_width] * block); output_shape.set(idx_height, input_shape[idx_height] * block); output_shape.set(idx_channel, input_shape[idx_channel] / (block * block)); @@ -1173,10 +1228,10 @@ inline TensorShape compute_split_shape(const ITensorInfo *input, unsigned int ax TensorShape empty_shape; empty_shape.set(0, 0); - TensorShape out_shape{ input->tensor_shape() }; + TensorShape out_shape{input->tensor_shape()}; // Return empty shape if axis is invalid - if(axis > input->tensor_shape().num_dimensions()) + if (axis > input->tensor_shape().num_dimensions()) { return empty_shape; } @@ -1184,7 +1239,7 @@ inline TensorShape compute_split_shape(const ITensorInfo *input, unsigned int ax size_t axis_size = out_shape[axis]; // Return empty shape if num_split is not valid - if(axis_size % num_splits) + if (axis_size % num_splits) { return empty_shape; } @@ -1203,9 +1258,10 @@ inline TensorShape compute_split_shape(const ITensorInfo *input, unsigned int ax * * @return the calculated shape */ -inline TensorShape compute_space_to_batch_shape(const ITensorInfo *input, int block_x, int block_y, const Size2D &padding_left, const Size2D &padding_right) +inline TensorShape compute_space_to_batch_shape( + const ITensorInfo *input, int block_x, int block_y, const Size2D &padding_left, const Size2D &padding_right) { - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; const DataLayout data_layout = input->data_layout(); const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -1231,7 +1287,7 @@ inline TensorShape compute_space_to_batch_shape(const ITensorInfo *input, int bl */ inline TensorShape compute_space_to_depth_shape(const ITensorInfo *input, int32_t block_shape) { - TensorShape output_shape{ input->tensor_shape() }; + TensorShape output_shape{input->tensor_shape()}; const DataLayout data_layout = input->data_layout(); const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -1276,7 +1332,7 @@ inline TensorShape compute_prior_box_shape(const ITensorInfo &input, const Prior inline TensorShape compute_padded_shape(const TensorShape &input_shape, const PaddingList &padding) { TensorShape padded_shape = input_shape; - for(size_t dim = 0; dim < padding.size(); ++dim) + for (size_t dim = 0; dim < padding.size(); ++dim) { const auto &padding_pair = padding[dim]; const uint32_t shape_on_index = (padded_shape.num_dimensions() <= dim) ? 1 : input_shape[dim]; @@ -1295,7 +1351,7 @@ inline TensorShape compute_padded_shape(const TensorShape &input_shape, const Pa inline TensorShape compute_tiled_shape(const TensorShape &input_shape, const Multiples &multiples) { TensorShape tiled_shape = input_shape; - for(size_t dim = 0; dim < multiples.size(); ++dim) + for (size_t dim = 0; dim < multiples.size(); ++dim) { tiled_shape.set(dim, input_shape[dim] * multiples[dim]); } @@ -1312,9 +1368,9 @@ inline TensorShape compute_tiled_shape(const TensorShape &input_shape, const Mul */ inline TensorShape compute_reduced_shape(const TensorShape &input, unsigned int axis, bool keep_dims = true) { - TensorShape output_shape{ input }; + TensorShape output_shape{input}; - if(!keep_dims) + if (!keep_dims) { output_shape.remove_dimension(axis); } @@ -1407,14 +1463,14 @@ inline TensorShape calculate_concatenate_shape(const std::vector<T *> &input, si #if defined(ARM_COMPUTE_ASSERTS_ENABLED) // All dimensions must match except the axis one - for(unsigned int i = 0; i < MAX_DIMS; ++i) + for (unsigned int i = 0; i < MAX_DIMS; ++i) { - if(i == axis) + if (i == axis) { continue; } - for(const auto &tensor : input) + for (const auto &tensor : input) { ARM_COMPUTE_ERROR_ON(tensor == nullptr); const TensorShape shape = extract_shape(tensor); @@ -1425,7 +1481,7 @@ inline TensorShape calculate_concatenate_shape(const std::vector<T *> &input, si // Calculate output shape size_t new_size = 0; - for(const auto &tensor : input) + for (const auto &tensor : input) { const TensorShape shape = extract_shape(tensor); new_size += shape[axis]; @@ -1448,14 +1504,14 @@ inline TensorShape compute_stack_shape(const ITensorInfo &a, unsigned int axis, ARM_COMPUTE_ERROR_ON(axis > a.num_dimensions()); ARM_COMPUTE_ERROR_ON(a.num_dimensions() > 4); - TensorShape shape_out{ a.tensor_shape() }; + TensorShape shape_out{a.tensor_shape()}; shape_out.set(axis, num_tensors); unsigned int i_shift = 0; - for(unsigned int i = 0; i < a.num_dimensions(); ++i) + for (unsigned int i = 0; i < a.num_dimensions(); ++i) { - if(i == axis) + if (i == axis) { i_shift++; } @@ -1473,7 +1529,8 @@ inline TensorShape compute_stack_shape(const ITensorInfo &a, unsigned int axis, * * @return the calculated shape */ -inline TensorShape compute_conv3d_shape(const TensorShape &src, const TensorShape &weights, const Conv3dInfo &conv3d_info) +inline TensorShape +compute_conv3d_shape(const TensorShape &src, const TensorShape &weights, const Conv3dInfo &conv3d_info) { // Weight tensor shape indices (D H W Cin Cout) constexpr unsigned int weights_depth_dim = 4u; @@ -1488,7 +1545,7 @@ inline TensorShape compute_conv3d_shape(const TensorShape &src, const TensorShap constexpr unsigned int width_dim = 1u; constexpr unsigned int channel_dim = 0u; - TensorShape output_shape{ src }; + TensorShape output_shape{src}; const size_t pad_left = conv3d_info.padding.left; const size_t pad_right = conv3d_info.padding.right; const size_t pad_top = conv3d_info.padding.top; @@ -1506,17 +1563,41 @@ inline TensorShape compute_conv3d_shape(const TensorShape &src, const TensorShap int output_height_size = 0; int output_depth_size = 0; - switch(conv3d_info.round_type) + switch (conv3d_info.round_type) { case DimensionRoundingType::FLOOR: - output_width_size = static_cast<int>(std::floor((static_cast<float>(src[width_dim] + pad_left + pad_right - (dilation_x * (weights[weights_width_dim] - 1) + 1)) / stride_x) + 1)); - output_height_size = static_cast<int>(std::floor((static_cast<float>(src[height_dim] + pad_top + pad_bottom - (dilation_y * (weights[weights_height_dim] - 1) + 1)) / stride_y) + 1)); - output_depth_size = static_cast<int>(std::floor((static_cast<float>(src[depth_dim] + pad_front + pad_back - (dilation_z * (weights[weights_depth_dim] - 1) + 1)) / stride_z) + 1)); + output_width_size = + static_cast<int>(std::floor((static_cast<float>(src[width_dim] + pad_left + pad_right - + (dilation_x * (weights[weights_width_dim] - 1) + 1)) / + stride_x) + + 1)); + output_height_size = + static_cast<int>(std::floor((static_cast<float>(src[height_dim] + pad_top + pad_bottom - + (dilation_y * (weights[weights_height_dim] - 1) + 1)) / + stride_y) + + 1)); + output_depth_size = + static_cast<int>(std::floor((static_cast<float>(src[depth_dim] + pad_front + pad_back - + (dilation_z * (weights[weights_depth_dim] - 1) + 1)) / + stride_z) + + 1)); break; case DimensionRoundingType::CEIL: - output_width_size = static_cast<int>(std::ceil((static_cast<float>(src[width_dim] + pad_left + pad_right - (dilation_x * (weights[weights_width_dim] - 1) + 1)) / stride_x) + 1)); - output_height_size = static_cast<int>(std::ceil((static_cast<float>(src[height_dim] + pad_top + pad_bottom - (dilation_y * (weights[weights_height_dim] - 1) + 1)) / stride_y) + 1)); - output_depth_size = static_cast<int>(std::ceil((static_cast<float>(src[depth_dim] + pad_front + pad_back - (dilation_z * (weights[weights_depth_dim] - 1) + 1)) / stride_z) + 1)); + output_width_size = + static_cast<int>(std::ceil((static_cast<float>(src[width_dim] + pad_left + pad_right - + (dilation_x * (weights[weights_width_dim] - 1) + 1)) / + stride_x) + + 1)); + output_height_size = + static_cast<int>(std::ceil((static_cast<float>(src[height_dim] + pad_top + pad_bottom - + (dilation_y * (weights[weights_height_dim] - 1) + 1)) / + stride_y) + + 1)); + output_depth_size = + static_cast<int>(std::ceil((static_cast<float>(src[depth_dim] + pad_front + pad_back - + (dilation_z * (weights[weights_depth_dim] - 1) + 1)) / + stride_z) + + 1)); break; default: ARM_COMPUTE_ERROR("Unsupported rounding type"); @@ -1539,7 +1620,7 @@ inline TensorShape compute_conv3d_shape(const TensorShape &src, const TensorShap */ inline TensorShape compute_pool3d_shape(const TensorShape &src, Pooling3dLayerInfo pool3d_info) { - TensorShape output_shape{ src }; + TensorShape output_shape{src}; const auto data_layout = DataLayout::NDHWC; const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); @@ -1552,10 +1633,12 @@ inline TensorShape compute_pool3d_shape(const TensorShape &src, Pooling3dLayerIn int output_height = 0; int output_depth = 0; - std::tie(output_width, output_height, output_depth) = scaled_3d_dimensions_signed(src[idx_width], src[idx_height], src[idx_depth], pool_size_width, pool_size_height, - pool_size_depth, pool3d_info); + std::tie(output_width, output_height, output_depth) = + scaled_3d_dimensions_signed(src[idx_width], src[idx_height], src[idx_depth], pool_size_width, pool_size_height, + pool_size_depth, pool3d_info); - ARM_COMPUTE_ERROR_ON_MSG((output_width < 1 || output_height < 1 || output_depth < 1), "Calculated output dimension size is invalid"); + ARM_COMPUTE_ERROR_ON_MSG((output_width < 1 || output_height < 1 || output_depth < 1), + "Calculated output dimension size is invalid"); output_shape.set(idx_width, static_cast<size_t>(output_width)); output_shape.set(idx_height, static_cast<size_t>(output_height)); @@ -1576,7 +1659,8 @@ inline TensorShape compute_pool3d_shape(const TensorShape &src, Pooling3dLayerIn * * @return the calculated shape */ -inline TensorShape compute_gather_shape(const TensorShape &input_shape, const TensorShape &indices_shape, uint32_t actual_axis) +inline TensorShape +compute_gather_shape(const TensorShape &input_shape, const TensorShape &indices_shape, uint32_t actual_axis) { const auto input_num_dims = input_shape.num_dimensions(); const auto indices_num_dims = indices_shape.num_dimensions(); @@ -1587,22 +1671,23 @@ inline TensorShape compute_gather_shape(const TensorShape &input_shape, const Te TensorShape output_shape; size_t dim_no = 0; - for(; dim_no < actual_axis; ++dim_no) + for (; dim_no < actual_axis; ++dim_no) { output_shape.set(dim_no, input_shape[dim_no]); } - for(; dim_no < actual_axis + indices_num_dims; ++dim_no) + for (; dim_no < actual_axis + indices_num_dims; ++dim_no) { output_shape.set(dim_no, indices_shape[dim_no - actual_axis]); } - for(; dim_no < input_num_dims + indices_num_dims - 1; ++dim_no) + for (; dim_no < input_num_dims + indices_num_dims - 1; ++dim_no) { output_shape.set(dim_no, input_shape[dim_no + 1 - indices_num_dims]); } - ARM_COMPUTE_ERROR_ON(input_shape.total_size() * indices_shape.total_size() != output_shape.total_size() * input_shape[actual_axis]); + ARM_COMPUTE_ERROR_ON(input_shape.total_size() * indices_shape.total_size() != + output_shape.total_size() * input_shape[actual_axis]); return output_shape; } diff --git a/arm_compute/core/utils/misc/Traits.h b/arm_compute/core/utils/misc/Traits.h index 933922f63c..944fcb95f9 100644 --- a/arm_compute/core/utils/misc/Traits.h +++ b/arm_compute/core/utils/misc/Traits.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_UTILS_TRAITS_TRAITS_H #include "arm_compute/core/Types.h" + #include <type_traits> namespace arm_compute diff --git a/arm_compute/core/utils/misc/Utility.h b/arm_compute/core/utils/misc/Utility.h index e3e20d719f..22f10d74cc 100644 --- a/arm_compute/core/utils/misc/Utility.h +++ b/arm_compute/core/utils/misc/Utility.h @@ -44,7 +44,7 @@ struct index_sequence }; template <std::size_t N, std::size_t... S> -struct index_sequence_generator : index_sequence_generator < N - 1, N - 1, S... > +struct index_sequence_generator : index_sequence_generator<N - 1, N - 1, S...> { }; @@ -58,17 +58,17 @@ template <std::size_t N> using index_sequence_t = typename index_sequence_generator<N>::type; template <typename T, std::size_t N, T val, T... vals> -struct generate_array : generate_array < T, N - 1, val, val, vals... > +struct generate_array : generate_array<T, N - 1, val, val, vals...> { }; template <typename T, T val, T... vals> struct generate_array<T, 0, val, vals...> { - static constexpr std::array<T, sizeof...(vals)> value{ vals... }; + static constexpr std::array<T, sizeof...(vals)> value{vals...}; }; -template <typename T, T val, T... vals> +template <typename T, T val, T... vals> constexpr std::array<T, sizeof...(vals)> generate_array<T, 0, val, vals...>::value; /** @endcond */ @@ -79,7 +79,7 @@ template <std::size_t... S, typename T = std::array<typename std::iterator_traits<Iterator>::value_type, sizeof...(S)>> T make_array(Iterator first, index_sequence<S...>) { - return T{ { first[S]... } }; + return T{{first[S]...}}; } } // namespace detail @@ -87,7 +87,7 @@ template <std::size_t N, typename Iterator> std::array<typename std::iterator_traits<Iterator>::value_type, N> make_array(Iterator first, Iterator last) { ARM_COMPUTE_UNUSED(last); - return detail::make_array(first, index_sequence_t<N> {}); + return detail::make_array(first, index_sequence_t<N>{}); } /** Performs clamping among a lower and upper value. @@ -119,7 +119,7 @@ inline void for_each(F &&) * @param[in] args Remaining arguments */ template <typename F, typename T, typename... Ts> -inline void for_each(F &&func, T &&arg, Ts &&... args) +inline void for_each(F &&func, T &&arg, Ts &&...args) { func(std::forward<T>(arg)); for_each(std::forward<F>(func), std::forward<Ts>(args)...); @@ -143,9 +143,11 @@ inline T &&foldl(F &&, T &&value) * @param[in] values Remaining arguments */ template <typename F, typename T, typename U, typename... Us> -inline auto foldl(F &&func, T &&initial, U &&value, Us &&... values) -> decltype(func(std::forward<T>(initial), std::forward<U>(value))) +inline auto foldl(F &&func, T &&initial, U &&value, Us &&...values) + -> decltype(func(std::forward<T>(initial), std::forward<U>(value))) { - return foldl(std::forward<F>(func), func(std::forward<T>(initial), std::forward<U>(value)), std::forward<Us>(values)...); + return foldl(std::forward<F>(func), func(std::forward<T>(initial), std::forward<U>(value)), + std::forward<Us>(values)...); } /** Perform an index sort of a given vector. @@ -160,11 +162,7 @@ std::vector<size_t> sort_indices(const std::vector<T> &v) std::vector<size_t> idx(v.size()); std::iota(idx.begin(), idx.end(), 0); - std::sort(idx.begin(), idx.end(), - [&v](size_t i1, size_t i2) - { - return v[i1] < v[i2]; - }); + std::sort(idx.begin(), idx.end(), [&v](size_t i1, size_t i2) { return v[i1] < v[i2]; }); return idx; } @@ -178,7 +176,7 @@ std::vector<size_t> sort_indices(const std::vector<T> &v) */ inline bool endswith(const std::string &str, const std::string &suffix) { - if(str.size() < suffix.size()) + if (str.size() < suffix.size()) { return false; } @@ -205,10 +203,7 @@ inline bool check_aligned(void *ptr, const size_t alignment) */ inline std::string tolower(std::string string) { - std::transform(string.begin(), string.end(), string.begin(), [](unsigned char c) - { - return std::tolower(c); - }); + std::transform(string.begin(), string.end(), string.begin(), [](unsigned char c) { return std::tolower(c); }); return string; } @@ -227,7 +222,7 @@ inline std::string getenv(const std::string &env_name) return std::string{}; #else // BARE_METAL const auto env_chr = std::getenv(env_name.c_str()); - return env_chr == nullptr ? std::string{} : std::string{ env_chr }; + return env_chr == nullptr ? std::string{} : std::string{env_chr}; #endif // BARE_METAL } } // namespace utility diff --git a/arm_compute/core/utils/quantization/AsymmHelpers.h b/arm_compute/core/utils/quantization/AsymmHelpers.h index a15f3e5cde..2324fe1838 100644 --- a/arm_compute/core/utils/quantization/AsymmHelpers.h +++ b/arm_compute/core/utils/quantization/AsymmHelpers.h @@ -41,7 +41,10 @@ namespace quantization * * @return a status */ -Status calculate_quantized_multiplier(float multiplier, int32_t *quant_multiplier, int32_t *shift, bool ignore_epsilon = false); +Status calculate_quantized_multiplier(float multiplier, + int32_t *quant_multiplier, + int32_t *shift, + bool ignore_epsilon = false); /** Calculate quantized representation of multiplier with value less than one. * * @param[in] multiplier Real multiplier. @@ -51,7 +54,10 @@ Status calculate_quantized_multiplier(float multiplier, int32_t *quant_multiplie * * @return a status */ -Status calculate_quantized_multiplier_less_than_one(float multiplier, int32_t *quant_multiplier, int32_t *right_shift, bool ignore_epsilon = false); +Status calculate_quantized_multiplier_less_than_one(float multiplier, + int32_t *quant_multiplier, + int32_t *right_shift, + bool ignore_epsilon = false); /** Calculate quantized representation of multiplier having value greater than one. * * @param[in] multiplier Real multiplier. @@ -60,7 +66,8 @@ Status calculate_quantized_multiplier_less_than_one(float multiplier, int32_t *q * * @return a status */ -Status calculate_quantized_multiplier_greater_than_one(float multiplier, int32_t *quantized_multiplier, int32_t *left_shift); +Status +calculate_quantized_multiplier_greater_than_one(float multiplier, int32_t *quantized_multiplier, int32_t *left_shift); /** Calculate quantized representation of per-channel multipliers * @@ -71,9 +78,9 @@ Status calculate_quantized_multiplier_greater_than_one(float multiplier, int32_t * * @return a status */ -Status calculate_quantized_multipliers(const QuantizationInfo &iq_info, - const QuantizationInfo &wq_info, - const QuantizationInfo &oq_info, +Status calculate_quantized_multipliers(const QuantizationInfo &iq_info, + const QuantizationInfo &wq_info, + const QuantizationInfo &oq_info, GEMMLowpOutputStageInfo &stage_info); /** Get minimum and maximum values for the input quantized data type @@ -147,7 +154,10 @@ int32_t saturating_rounding_multiply_by_pow2(int32_t exponent, int32_t v); * @param[out] output_shift Shift for inverse square root * */ -void get_invsqrt_quantized_multiplier_exp(int32_t input, int32_t reverse_shift, int32_t &output_inv_sqrt, int32_t &output_shift); +void get_invsqrt_quantized_multiplier_exp(int32_t input, + int32_t reverse_shift, + int32_t &output_inv_sqrt, + int32_t &output_shift); } // namespace quantization } // namespace arm_compute diff --git a/arm_compute/dynamic_fusion/runtime/gpu/cl/ClWorkloadRuntime.h b/arm_compute/dynamic_fusion/runtime/gpu/cl/ClWorkloadRuntime.h index b5af589cd2..3deaff74fc 100644 --- a/arm_compute/dynamic_fusion/runtime/gpu/cl/ClWorkloadRuntime.h +++ b/arm_compute/dynamic_fusion/runtime/gpu/cl/ClWorkloadRuntime.h @@ -26,6 +26,7 @@ #include "arm_compute/core/TensorInfo.h" #include "arm_compute/dynamic_fusion/sketch/MemoryDescriptor.h" + #include <map> #include <memory> diff --git a/arm_compute/dynamic_fusion/sketch/MemoryDescriptor.h b/arm_compute/dynamic_fusion/sketch/MemoryDescriptor.h index 3daedd4efb..e2a34e4424 100644 --- a/arm_compute/dynamic_fusion/sketch/MemoryDescriptor.h +++ b/arm_compute/dynamic_fusion/sketch/MemoryDescriptor.h @@ -53,10 +53,12 @@ namespace dynamic_fusion enum class MemoryType { /** Both User and Auxiliary types are of Alloc type. Since they require memory allocation */ - User = 0, /**< Memory coming directly from users, e.g. for argument tensors */ - Auxiliary = 1, /**< Additional memory required by the workload tensor, e.g. for tensors holding temporary results between kernels */ + User = 0, /**< Memory coming directly from users, e.g. for argument tensors */ + Auxiliary = + 1, /**< Additional memory required by the workload tensor, e.g. for tensors holding temporary results between kernels */ /** Virtual type is of No-Alloc type. Since it doesn't require memory allocation */ - Virtual = 2, /**< Temporary tile which is not allocated as a whole tensor in the memory. It is mainly used at sketch time to link operators; there should be no Virtual tensors at runtime */ + Virtual = + 2, /**< Temporary tile which is not allocated as a whole tensor in the memory. It is mainly used at sketch time to link operators; there should be no Virtual tensors at runtime */ }; /** Memory information for tensors with @ref MemoryType::Auxiliary. @@ -66,9 +68,7 @@ struct AuxMemoryInfo { AuxMemoryInfo() = default; - AuxMemoryInfo(size_t size, size_t alignment = 0) noexcept - : size(size), - alignment(alignment) + AuxMemoryInfo(size_t size, size_t alignment = 0) noexcept : size(size), alignment(alignment) { } @@ -76,8 +76,8 @@ struct AuxMemoryInfo { return info0.size == info1.size && info0.alignment == info1.alignment; } - size_t size{ 0 }; /**< Total memory size in bytes */ - size_t alignment{ 0 }; /**< Memory alignment in bytes */ + size_t size{0}; /**< Total memory size in bytes */ + size_t alignment{0}; /**< Memory alignment in bytes */ }; /** Descriptor of a workload tensor memory */ diff --git a/arm_compute/dynamic_fusion/sketch/attributes/CastAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/CastAttributes.h index 59efc8bd5d..ba2f658c7c 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/CastAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/CastAttributes.h @@ -49,8 +49,8 @@ public: ConvertPolicy convert_policy() const; private: - DataType _data_type{}; /**< Data Type to be casted to */ - ConvertPolicy _convert_policy{ ConvertPolicy::SATURATE }; /**< Convert Policy */ + DataType _data_type{}; /**< Data Type to be casted to */ + ConvertPolicy _convert_policy{ConvertPolicy::SATURATE}; /**< Convert Policy */ }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/attributes/Conv2dAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/Conv2dAttributes.h index 58102d8d88..a98ef0363e 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/Conv2dAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/Conv2dAttributes.h @@ -26,6 +26,7 @@ #include "arm_compute/core/Size2D.h" #include "arm_compute/core/Types.h" + #include <cstdint> namespace arm_compute @@ -55,9 +56,9 @@ public: Size2D dilation() const; private: - Padding2D _pad{}; /**< Padding */ - Size2D _stride{ 1U, 1U }; /**< Stride */ - Size2D _dilation{ 1U, 1U }; /**< Dilation */ + Padding2D _pad{}; /**< Padding */ + Size2D _stride{1U, 1U}; /**< Stride */ + Size2D _dilation{1U, 1U}; /**< Dilation */ }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/attributes/DepthwiseConv2dAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/DepthwiseConv2dAttributes.h index 6d05e9e4d6..c46b25cb5d 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/DepthwiseConv2dAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/DepthwiseConv2dAttributes.h @@ -26,6 +26,7 @@ #include "arm_compute/core/Size2D.h" #include "arm_compute/core/Types.h" + #include <cstdint> namespace arm_compute @@ -63,11 +64,11 @@ public: DimensionRoundingType dimension_rounding_type() const; private: - Padding2D _pad{}; /**< Padding */ - Size2D _stride{ 1U, 1U }; /**< Stride */ - Size2D _dilation{ 1U, 1U }; /**< Dilation */ - uint32_t _depth_multiplier{ 1U }; /**< Depth multiplier */ - DimensionRoundingType _dimension_rounding_type{ DimensionRoundingType::FLOOR }; /**< Dimension rounding type */ + Padding2D _pad{}; /**< Padding */ + Size2D _stride{1U, 1U}; /**< Stride */ + Size2D _dilation{1U, 1U}; /**< Dilation */ + uint32_t _depth_multiplier{1U}; /**< Depth multiplier */ + DimensionRoundingType _dimension_rounding_type{DimensionRoundingType::FLOOR}; /**< Dimension rounding type */ }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/attributes/Pool2dAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/Pool2dAttributes.h index be30781d86..19d8b96dcf 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/Pool2dAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/Pool2dAttributes.h @@ -72,8 +72,8 @@ private: PoolingType _pool_type{}; Padding2D _pad{}; Size2D _pool_size{}; - Size2D _stride{ 1U, 1U }; - bool _exclude_padding{ true }; + Size2D _stride{1U, 1U}; + bool _exclude_padding{true}; }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/attributes/ResizeAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/ResizeAttributes.h index 8992693cd1..7410cc7e70 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/ResizeAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/ResizeAttributes.h @@ -27,6 +27,7 @@ #include "arm_compute/core/PixelValue.h" #include "arm_compute/core/Types.h" + #include <cstdint> namespace arm_compute @@ -75,9 +76,9 @@ public: private: int32_t _output_width{}; int32_t _output_height{}; - InterpolationPolicy _interpolation_policy{ InterpolationPolicy::BILINEAR }; - SamplingPolicy _sampling_policy{ SamplingPolicy::CENTER }; - bool _align_corners{ false }; + InterpolationPolicy _interpolation_policy{InterpolationPolicy::BILINEAR}; + SamplingPolicy _sampling_policy{SamplingPolicy::CENTER}; + bool _align_corners{false}; }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/attributes/SoftmaxAttributes.h b/arm_compute/dynamic_fusion/sketch/attributes/SoftmaxAttributes.h index fc50aa0f68..61654bfa27 100644 --- a/arm_compute/dynamic_fusion/sketch/attributes/SoftmaxAttributes.h +++ b/arm_compute/dynamic_fusion/sketch/attributes/SoftmaxAttributes.h @@ -51,7 +51,7 @@ public: int axis() const; private: - float _beta{ 1.f }; /**< Scaling factor for the exponent */ + float _beta{1.f}; /**< Scaling factor for the exponent */ bool _is_log_softmax{}; /**< True if operation is log-softmax */ int _axis{}; /**< Axis/Dimension to perform the operation */ }; diff --git a/arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadContext.h b/arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadContext.h index 0b60899734..38b350c7eb 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadContext.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/GpuWorkloadContext.h @@ -85,7 +85,7 @@ public: * @return TensorInfo Newly created tensor info */ template <typename... TArgs> - TensorInfo create_tensor_info(TArgs &&... args) + TensorInfo create_tensor_info(TArgs &&...args) { auto tensor_info = TensorInfo(std::forward<TArgs>(args)...); register_user_tensor(tensor_info); diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuAdd.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuAdd.h index 33eded4dff..5b6c1b90ab 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuAdd.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuAdd.h @@ -65,9 +65,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *lhs, - ITensorInfo *rhs); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *lhs, ITensorInfo *rhs); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running @@ -76,18 +74,14 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *lhs, - const ITensorInfo *rhs); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *lhs, const ITensorInfo *rhs); /** Validate the operator and check if the its configuration is supported and if it can be fused into the workload sketch. * * Parameters are similar to @ref GpuAdd::create_op() * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *rhs, - const ITensorInfo *lhs); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *rhs, const ITensorInfo *lhs); }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuCast.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuCast.h index 83b004b8b8..1593cec804 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuCast.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuCast.h @@ -68,9 +68,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - const Attributes &attributes); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running @@ -79,18 +77,15 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src, - const Attributes &attributes); + static Status + is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src, const Attributes &attributes); /** Validate the operator and check if the its configuration is supported and if it can be fused into the workload sketch. * * Parameters are similar to @ref GpuCast::create_op() * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src, - const Attributes &attributes); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src, const Attributes &attributes); }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuClamp.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuClamp.h index e96251196a..5dd77bdc8e 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuClamp.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuClamp.h @@ -62,9 +62,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - const Attributes &attributes); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * @@ -74,9 +72,8 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src, - const Attributes &attributes); + static Status + is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src, const Attributes &attributes); /** Validate the operator and check if it can be fused into the workload sketch. * @@ -84,9 +81,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src, - const Attributes &attributes); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src, const Attributes &attributes); }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuConv2d.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuConv2d.h index 612cc83a1f..da7e860757 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuConv2d.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuConv2d.h @@ -64,11 +64,8 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - ITensorInfo *wei, - ITensorInfo *bia, - const Attributes &attributes); + static ITensorInfo *create_op( + GpuWorkloadSketch &sketch, ITensorInfo *src, ITensorInfo *wei, ITensorInfo *bia, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuDepthwiseConv2d.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuDepthwiseConv2d.h index a0cb292730..958569efd7 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuDepthwiseConv2d.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuDepthwiseConv2d.h @@ -63,11 +63,8 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - ITensorInfo *wei, - ITensorInfo *bia, - const Attributes &attributes); + static ITensorInfo *create_op( + GpuWorkloadSketch &sketch, ITensorInfo *src, ITensorInfo *wei, ITensorInfo *bia, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuMul.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuMul.h index 3e0ebdd96c..d13e4a3cad 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuMul.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuMul.h @@ -62,9 +62,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *lhs, - ITensorInfo *rhs); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *lhs, ITensorInfo *rhs); /** Check if the operator configuration is supported, irrespective of fusion * @@ -74,9 +72,7 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *lhs, - const ITensorInfo *rhs); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *lhs, const ITensorInfo *rhs); /** Validate the operator and check if the configuration is supported and if it can be fused into the workload sketch. * @@ -84,9 +80,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *rhs, - const ITensorInfo *lhs); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *rhs, const ITensorInfo *lhs); }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h index 06317511cd..deb5559b9d 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h @@ -56,9 +56,7 @@ public: * @param[in, out] dst Destination tensor info. * If an uninitialized ITensorInfo is passed in, it will be auto-initialized. */ - static void create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - ITensorInfo *dst); + static void create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, ITensorInfo *dst); /** Check if the operator configuration is supported, irrespective of fusion. * @@ -68,9 +66,7 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src, - const ITensorInfo *dst); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src, const ITensorInfo *dst); /** Validate the operator and check if the its configuration is supported and if it can be fused into the workload sketch. * @@ -78,9 +74,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src, - const ITensorInfo *dst); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src, const ITensorInfo *dst); }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuPool2d.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuPool2d.h index 65a092c0a2..4d2db0e89c 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuPool2d.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuPool2d.h @@ -55,8 +55,8 @@ public: GpuPool2dSettings use_inf_as_limit(bool use_inf_as_limit); private: - bool _mixed_precision{ false }; - bool _use_inf_as_limit{ true }; + bool _mixed_precision{false}; + bool _use_inf_as_limit{true}; }; /** Operator interface. */ @@ -86,10 +86,8 @@ public: * @param[in] attributes Operator attributes * @param[in] settings Operator settings */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - const Attributes &attributes, - const Settings &settings); + static ITensorInfo * + create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, const Attributes &attributes, const Settings &settings); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuReshape.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuReshape.h index 0f50127199..dc194fcadb 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuReshape.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuReshape.h @@ -62,9 +62,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - const Attributes &attributes); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running @@ -73,18 +71,15 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src, - const Attributes &attributes); + static Status + is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src, const Attributes &attributes); /** Validate the operator and check if the its configuration is supported and if it can be fused into the workload sketch. * * Parameters are similar to @ref GpuReshape::create_op() * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src, - const Attributes &attributes); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src, const Attributes &attributes); }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuResize.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuResize.h index 2579d10f5b..e2ece80a1d 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuResize.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuResize.h @@ -67,9 +67,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - const Attributes &attributes); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running @@ -78,18 +76,15 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src, - const Attributes &attributes); + static Status + is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src, const Attributes &attributes); /** Validate the operator and check if the its configuration is supported and if it can be fused into the workload sketch. * * Parameters are similar to @ref GpuResize::create_op() * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src, - const Attributes &attributes); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src, const Attributes &attributes); }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSigmoid.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSigmoid.h index 616a61e614..798b84b906 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSigmoid.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSigmoid.h @@ -59,8 +59,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src); /** Check if the operator configuration is supported, irrespective of fusion * @@ -69,8 +68,7 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src); /** Validate the operator and check if it can be fused into the workload sketch. * @@ -78,8 +76,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src); }; } // namespace dynamic_fusion diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSoftmax.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSoftmax.h index e86ef91e6a..66c2d77310 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSoftmax.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSoftmax.h @@ -62,10 +62,7 @@ public: * @param[in] dst Destination tensor info. * @param[in] attributes Operator attributes */ - static void create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src, - ITensorInfo *dst, - const Attributes &attributes); + static void create_op(GpuWorkloadSketch &sketch, ITensorInfo *src, ITensorInfo *dst, const Attributes &attributes); /** Check if the operator configuration is supported, irrespective of fusion * * @param[in] context Workload context within which the operator is running diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSub.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSub.h index 6f8c2d0b76..2d9255fff2 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSub.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuSub.h @@ -65,9 +65,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *lhs, - ITensorInfo *rhs); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *lhs, ITensorInfo *rhs); /** Check if the operator configuration is supported, irrespective of fusion * @@ -77,9 +75,7 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *lhs, - const ITensorInfo *rhs); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *lhs, const ITensorInfo *rhs); /** Validate the operator and check if its configuration is supported and if it can be fused into the workload sketch. * @@ -87,9 +83,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *rhs, - const ITensorInfo *lhs); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *rhs, const ITensorInfo *lhs); }; } // namespace dynamic_fusion } // namespace experimental diff --git a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuTanh.h b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuTanh.h index 08b5032e93..9c0ce6de02 100644 --- a/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuTanh.h +++ b/arm_compute/dynamic_fusion/sketch/gpu/operators/GpuTanh.h @@ -59,8 +59,7 @@ public: * * @return Pointer for the destination tensor info */ - static ITensorInfo *create_op(GpuWorkloadSketch &sketch, - ITensorInfo *src); + static ITensorInfo *create_op(GpuWorkloadSketch &sketch, ITensorInfo *src); /** Check if the operator configuration is supported, irrespective of fusion * @@ -69,8 +68,7 @@ public: * * @return Status */ - static Status is_supported_op(const GpuWorkloadContext &context, - const ITensorInfo *src); + static Status is_supported_op(const GpuWorkloadContext &context, const ITensorInfo *src); /** Validate the operator and check if it can be fused into the workload sketch. * @@ -78,8 +76,7 @@ public: * * @return Status */ - static Status validate_op(const GpuWorkloadSketch &sketch, - const ITensorInfo *src); + static Status validate_op(const GpuWorkloadSketch &sketch, const ITensorInfo *src); }; } // namespace dynamic_fusion diff --git a/arm_compute/function_info/ActivationLayerInfo.h b/arm_compute/function_info/ActivationLayerInfo.h index 84e962cb3a..195b67cf99 100644 --- a/arm_compute/function_info/ActivationLayerInfo.h +++ b/arm_compute/function_info/ActivationLayerInfo.h @@ -39,17 +39,17 @@ enum class ActivationFunction RELU, /**< Rectifier ( \f$ f(x) = max(0,x) \f$ ) */ BOUNDED_RELU, /**< Upper Bounded Rectifier ( \f$ f(x) = min(a, max(0,x)) \f$ ) */ LU_BOUNDED_RELU, /**< Lower and Upper Bounded Rectifier ( \f$ f(x) = min(a, max(b,x)) \f$ ) */ - LEAKY_RELU, /**< Leaky Rectifier ( \f$ f(x) = \begin{cases} \alpha x & \quad \text{if } x \text{ < 0}\\ x & \quad \text{if } x \geq \text{ 0 } \end{cases} \f$ ) */ - SOFT_RELU, /**< Soft Rectifier ( \f$ f(x)= log(1+e^x) \f$ ) */ - ELU, /**< Exponential Linear Unit ( \f$ f(x) = \begin{cases} \alpha (exp(x) - 1) & \quad \text{if } x \text{ < 0}\\ x & \quad \text{if } x \geq \text{ 0 } \end{cases} \f$ ) */ - ABS, /**< Absolute ( \f$ f(x)= |x| \f$ ) */ - SQUARE, /**< Square ( \f$ f(x)= x^2 \f$ )*/ - SQRT, /**< Square root ( \f$ f(x) = \sqrt{x} \f$ )*/ - LINEAR, /**< Linear ( \f$ f(x)= ax + b \f$ ) */ - IDENTITY, /**< Identity ( \f$ f(x)= x \f$ ) */ - HARD_SWISH, /**< Hard-swish ( \f$ f(x) = (x \text{ReLU6}(x+3))/6 = x \min(\max(0,x+3),6)/6 \f$ ) */ - SWISH, /**< Swish ( \f$ f(x) = \frac{x}{1 + e^{-ax}} = x \text{logistic}(ax) \f$ ) */ - GELU /**< GELU ( \f$ f(x) = x * 1/2 * 1 + erf(x / \sqrt{2}) \f$ ) */ + LEAKY_RELU, /**< Leaky Rectifier ( \f$ f(x) = \begin{cases} \alpha x & \quad \text{if } x \text{ < 0}\\ x & \quad \text{if } x \geq \text{ 0 } \end{cases} \f$ ) */ + SOFT_RELU, /**< Soft Rectifier ( \f$ f(x)= log(1+e^x) \f$ ) */ + ELU, /**< Exponential Linear Unit ( \f$ f(x) = \begin{cases} \alpha (exp(x) - 1) & \quad \text{if } x \text{ < 0}\\ x & \quad \text{if } x \geq \text{ 0 } \end{cases} \f$ ) */ + ABS, /**< Absolute ( \f$ f(x)= |x| \f$ ) */ + SQUARE, /**< Square ( \f$ f(x)= x^2 \f$ )*/ + SQRT, /**< Square root ( \f$ f(x) = \sqrt{x} \f$ )*/ + LINEAR, /**< Linear ( \f$ f(x)= ax + b \f$ ) */ + IDENTITY, /**< Identity ( \f$ f(x)= x \f$ ) */ + HARD_SWISH, /**< Hard-swish ( \f$ f(x) = (x \text{ReLU6}(x+3))/6 = x \min(\max(0,x+3),6)/6 \f$ ) */ + SWISH, /**< Swish ( \f$ f(x) = \frac{x}{1 + e^{-ax}} = x \text{logistic}(ax) \f$ ) */ + GELU /**< GELU ( \f$ f(x) = x * 1/2 * 1 + erf(x / \sqrt{2}) \f$ ) */ }; /** Activation Layer Information class */ class ActivationLayerInfo @@ -68,8 +68,7 @@ public: * (@ref ActivationFunction::BOUNDED_RELU, @ref ActivationFunction::LU_BOUNDED_RELU, @ref ActivationFunction::LINEAR, @ref ActivationFunction::TANH). * @param[in] b (Optional) The beta parameter used by some activation functions (@ref ActivationFunction::LINEAR, @ref ActivationFunction::LU_BOUNDED_RELU, @ref ActivationFunction::TANH). */ - ActivationLayerInfo(ActivationFunction f, float a = 0.0f, float b = 0.0f) - : _act(f), _a(a), _b(b), _enabled(true) + ActivationLayerInfo(ActivationFunction f, float a = 0.0f, float b = 0.0f) : _act(f), _a(a), _b(b), _enabled(true) { } /** Get the type of activation function */ @@ -104,10 +103,10 @@ public: } #endif // __aarch64__ private: - ActivationFunction _act = { ActivationLayerInfo::ActivationFunction::IDENTITY }; + ActivationFunction _act = {ActivationLayerInfo::ActivationFunction::IDENTITY}; float _a = {}; float _b = {}; - bool _enabled = { false }; + bool _enabled = {false}; #ifdef __aarch64__ LookupTable256 _lut = {}; diff --git a/arm_compute/function_info/ConvolutionInfo.h b/arm_compute/function_info/ConvolutionInfo.h index c27dc523c8..4830cae137 100644 --- a/arm_compute/function_info/ConvolutionInfo.h +++ b/arm_compute/function_info/ConvolutionInfo.h @@ -33,14 +33,18 @@ namespace arm_compute struct ConvolutionInfo { ConvolutionInfo() = default; - ConvolutionInfo(const PadStrideInfo &pad_stride_info, unsigned int depth_multiplier, const ActivationLayerInfo &act_info, const Size2D &dilation) + ConvolutionInfo(const PadStrideInfo &pad_stride_info, + unsigned int depth_multiplier, + const ActivationLayerInfo &act_info, + const Size2D &dilation) : pad_stride_info(pad_stride_info), depth_multiplier(depth_multiplier), act_info(act_info), dilation(dilation) { } - PadStrideInfo pad_stride_info{}; /**< Convolution info (Pads, strides,...) */ - unsigned int depth_multiplier{ 1 }; /**< Multiplier to apply to input's depth to retrieve the output depth. Defaults to 1 */ - ActivationLayerInfo act_info{}; /**< Fused activation to apply after convolution. */ - Size2D dilation{ Size2D(1, 1) }; /**< Dilation, in elements, across x and y. Defaults to (1, 1). */ + PadStrideInfo pad_stride_info{}; /**< Convolution info (Pads, strides,...) */ + unsigned int depth_multiplier{ + 1}; /**< Multiplier to apply to input's depth to retrieve the output depth. Defaults to 1 */ + ActivationLayerInfo act_info{}; /**< Fused activation to apply after convolution. */ + Size2D dilation{Size2D(1, 1)}; /**< Dilation, in elements, across x and y. Defaults to (1, 1). */ }; } // namespace arm_compute #endif /* ACL_ARM_COMPUTE_FUNCTION_INFO_CONVOLUTIONINFO */ diff --git a/arm_compute/function_info/FullyConnectedLayerInfo.h b/arm_compute/function_info/FullyConnectedLayerInfo.h index 5f5578eadd..e65daeb2d4 100644 --- a/arm_compute/function_info/FullyConnectedLayerInfo.h +++ b/arm_compute/function_info/FullyConnectedLayerInfo.h @@ -35,13 +35,13 @@ struct FullyConnectedLayerInfo /* Fused-activation parameters */ ActivationLayerInfo activation_info{}; /**< Fused activation to apply after the matrix multiplication. */ /* Information about weights */ - DataLayout weights_trained_layout{ DataLayout::NCHW }; /**< Layout that the weights have been trained with. */ - bool transpose_weights{ true }; /**< Transpose weights if true. */ - bool are_weights_reshaped{ false }; /**< @deprecated Reshape the weights tensor if false. */ - bool retain_internal_weights{ false }; /**< Retain internal reshaped weights. */ - bool enable_fast_math{ false }; /**< Enable fast math computation. */ + DataLayout weights_trained_layout{DataLayout::NCHW}; /**< Layout that the weights have been trained with. */ + bool transpose_weights{true}; /**< Transpose weights if true. */ + bool are_weights_reshaped{false}; /**< @deprecated Reshape the weights tensor if false. */ + bool retain_internal_weights{false}; /**< Retain internal reshaped weights. */ + bool enable_fast_math{false}; /**< Enable fast math computation. */ /* Other parameters */ - bool fp_mixed_precision{ false }; /**< Use wider accumulators (32 bit instead of 16 for FP16) to improve accuracy. */ + bool fp_mixed_precision{false}; /**< Use wider accumulators (32 bit instead of 16 for FP16) to improve accuracy. */ /** Sets the weights trained data layout * diff --git a/arm_compute/function_info/GEMMInfo.h b/arm_compute/function_info/GEMMInfo.h index 29a57a00c2..c24762c0aa 100644 --- a/arm_compute/function_info/GEMMInfo.h +++ b/arm_compute/function_info/GEMMInfo.h @@ -26,6 +26,7 @@ #include "arm_compute/core/CoreTypes.h" #include "arm_compute/function_info/ActivationLayerInfo.h" + #include <vector> namespace arm_compute @@ -43,17 +44,22 @@ enum class GEMMLowpOutputStageType /** GEMMLowp output stage info */ struct GEMMLowpOutputStageInfo { - GEMMLowpOutputStageType type{ GEMMLowpOutputStageType::NONE }; /**< GEMMLowp output stage type */ - int32_t gemmlowp_offset{ 0 }; /**< GEMMLowp output stage offset used for quantizing to QASYMM8 */ - int32_t gemmlowp_multiplier{ 0 }; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ - int32_t gemmlowp_shift{ 0 }; /**< GEMMLowp output stage shift used for quantizing to uint8 */ - int32_t gemmlowp_min_bound{ std::numeric_limits<int32_t>::lowest() }; /**< GEMMLowp min value used to saturate down the output result before converting back to QASYMM8 */ - int32_t gemmlowp_max_bound{ std::numeric_limits<int32_t>::max() }; /**< GEMMLowp max value used to saturate down the output result before converting back to QASYMM8 */ - std::vector<int32_t> gemmlowp_multipliers{}; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ - std::vector<int32_t> gemmlowp_shifts{}; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ - float gemmlowp_real_multiplier{ 0 }; /**< GEMMLowp output stage real multiplier used for quantizing to QASYMM8 */ - bool is_quantized_per_channel{ false }; /**< GEMMLowp quantized per-channel flag */ - DataType output_data_type{ DataType::UNKNOWN }; /**< Output tensor data type to use if the output is not initialized */ + GEMMLowpOutputStageType type{GEMMLowpOutputStageType::NONE}; /**< GEMMLowp output stage type */ + int32_t gemmlowp_offset{0}; /**< GEMMLowp output stage offset used for quantizing to QASYMM8 */ + int32_t gemmlowp_multiplier{0}; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ + int32_t gemmlowp_shift{0}; /**< GEMMLowp output stage shift used for quantizing to uint8 */ + int32_t gemmlowp_min_bound{ + std::numeric_limits<int32_t>:: + lowest()}; /**< GEMMLowp min value used to saturate down the output result before converting back to QASYMM8 */ + int32_t gemmlowp_max_bound{ + std::numeric_limits<int32_t>:: + max()}; /**< GEMMLowp max value used to saturate down the output result before converting back to QASYMM8 */ + std::vector<int32_t> gemmlowp_multipliers{}; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ + std::vector<int32_t> gemmlowp_shifts{}; /**< GEMMLowp output stage multiplier used for quantizing to QASYMM8 */ + float gemmlowp_real_multiplier{0}; /**< GEMMLowp output stage real multiplier used for quantizing to QASYMM8 */ + bool is_quantized_per_channel{false}; /**< GEMMLowp quantized per-channel flag */ + DataType output_data_type{ + DataType::UNKNOWN}; /**< Output tensor data type to use if the output is not initialized */ }; /** GEMM information class. This class stores the necessary information to compute GEMM functions * @@ -100,9 +106,19 @@ public: * @param[in] fixed_format (Optional) Specify the selection of fixed format kernels for variable weights support in GEMM. These kernels expect the weights tensor to be in amemory format that is fixed by the kernel itself. For more information, see arm_compute::WeightFormat. * @param[in] weight_format (Optional) arm_gemm:WeightFormat enumeration requested by the user. Default is arm_compute::WeightFormat::UNSPECIFIED. */ - GEMMInfo(bool is_a_reshaped, bool is_b_reshaped, bool reshape_b_only_on_first_run, int depth_output_gemm3d = 0, bool reinterpret_input_as_3d = false, bool retain_internal_weights = false, - GEMMLowpOutputStageInfo gemmlowp_output_stage = GEMMLowpOutputStageInfo(), bool fp_mixed_precision = false, bool fast_math = false, bool broadcast_bias = false, - const ActivationLayerInfo &activation_info = ActivationLayerInfo(), bool fixed_format = false, arm_compute::WeightFormat weight_format = arm_compute::WeightFormat::UNSPECIFIED) noexcept + GEMMInfo(bool is_a_reshaped, + bool is_b_reshaped, + bool reshape_b_only_on_first_run, + int depth_output_gemm3d = 0, + bool reinterpret_input_as_3d = false, + bool retain_internal_weights = false, + GEMMLowpOutputStageInfo gemmlowp_output_stage = GEMMLowpOutputStageInfo(), + bool fp_mixed_precision = false, + bool fast_math = false, + bool broadcast_bias = false, + const ActivationLayerInfo &activation_info = ActivationLayerInfo(), + bool fixed_format = false, + arm_compute::WeightFormat weight_format = arm_compute::WeightFormat::UNSPECIFIED) noexcept : _is_a_reshaped(is_a_reshaped), _is_b_reshaped(is_b_reshaped), _reshape_b_only_on_first_run(reshape_b_only_on_first_run), diff --git a/arm_compute/function_info/MatMulInfo.h b/arm_compute/function_info/MatMulInfo.h index cd9ef1f4d9..fc73efb44a 100644 --- a/arm_compute/function_info/MatMulInfo.h +++ b/arm_compute/function_info/MatMulInfo.h @@ -55,8 +55,8 @@ public: } private: - bool _adj_lhs{ false }; - bool _adj_rhs{ false }; + bool _adj_lhs{false}; + bool _adj_rhs{false}; }; } // namespace arm_compute #endif /* ACL_ARM_COMPUTE_FUNCTION_INFO_MATMULINFO */ diff --git a/arm_compute/graph/Edge.h b/arm_compute/graph/Edge.h index 5e81b9c52f..7f5075d885 100644 --- a/arm_compute/graph/Edge.h +++ b/arm_compute/graph/Edge.h @@ -48,8 +48,18 @@ public: * @param[in] consumer_idx Consumer node input index * @param[in] tensor Tensor associated with the edge */ - Edge(EdgeID id, INode *producer, unsigned int producer_idx, INode *consumer, unsigned int consumer_idx, Tensor *tensor) - : _id(id), _producer(producer), _consumer(consumer), _producer_idx(producer_idx), _consumer_idx(consumer_idx), _tensor(tensor) + Edge(EdgeID id, + INode *producer, + unsigned int producer_idx, + INode *consumer, + unsigned int consumer_idx, + Tensor *tensor) + : _id(id), + _producer(producer), + _consumer(consumer), + _producer_idx(producer_idx), + _consumer_idx(consumer_idx), + _tensor(tensor) { } diff --git a/arm_compute/graph/Graph.h b/arm_compute/graph/Graph.h index 806d84c3fd..e6e173f5fa 100644 --- a/arm_compute/graph/Graph.h +++ b/arm_compute/graph/Graph.h @@ -79,7 +79,7 @@ public: * @return ID of the node */ template <typename NT, typename... Ts> - NodeID add_node(Ts &&... args); + NodeID add_node(Ts &&...args); /** Remove the node with the given ID * * @param[in] nid ID of the node to remove @@ -221,17 +221,17 @@ private: TensorID create_tensor(const TensorDescriptor &desc = TensorDescriptor()); private: - GraphID _id = GraphID(0); /**< Graph id */ - std::string _name = {}; /**< Graph name */ - std::vector<std::unique_ptr<INode>> _nodes = {}; /**< Graph nodes */ - std::vector<std::unique_ptr<Edge>> _edges = {}; /**< Graph edges */ - std::vector<std::unique_ptr<Tensor>> _tensors = {}; /**< Graph tensors */ + GraphID _id = GraphID(0); /**< Graph id */ + std::string _name = {}; /**< Graph name */ + std::vector<std::unique_ptr<INode>> _nodes = {}; /**< Graph nodes */ + std::vector<std::unique_ptr<Edge>> _edges = {}; /**< Graph edges */ + std::vector<std::unique_ptr<Tensor>> _tensors = {}; /**< Graph tensors */ std::map<NodeType, std::vector<NodeID>> _tagged_nodes = {}; /**< Graph nodes map with the node type as key */ - arm_compute::Mutex _mtx = {}; /**< Mutex used for graph construction */ + arm_compute::Mutex _mtx = {}; /**< Mutex used for graph construction */ }; template <typename NT, typename... Ts> -inline NodeID Graph::add_node(Ts &&... args) +inline NodeID Graph::add_node(Ts &&...args) { arm_compute::lock_guard<arm_compute::Mutex> lock(_mtx); @@ -245,7 +245,7 @@ inline NodeID Graph::add_node(Ts &&... args) _tagged_nodes[node->type()].push_back(nid); // Associate a new tensor with each output - for(auto &output : node->_outputs) + for (auto &output : node->_outputs) { output = create_tensor(); } diff --git a/arm_compute/graph/GraphBuilder.h b/arm_compute/graph/GraphBuilder.h index cb88c0e7aa..118d06bdda 100644 --- a/arm_compute/graph/GraphBuilder.h +++ b/arm_compute/graph/GraphBuilder.h @@ -51,7 +51,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_const_node(Graph &g, NodeParams params, const TensorDescriptor &desc, ITensorAccessorUPtr accessor = nullptr); + static NodeID + add_const_node(Graph &g, NodeParams params, const TensorDescriptor &desc, ITensorAccessorUPtr accessor = nullptr); /** Adds an input layer node to the graph * * @param[in] g Graph to add the node to @@ -61,7 +62,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_input_node(Graph &g, NodeParams params, const TensorDescriptor &desc, ITensorAccessorUPtr accessor = nullptr); + static NodeID + add_input_node(Graph &g, NodeParams params, const TensorDescriptor &desc, ITensorAccessorUPtr accessor = nullptr); /** Adds an output layer node to the graph * * @param[in] g Graph to add the node to @@ -71,7 +73,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_output_node(Graph &g, NodeParams params, NodeIdxPair input, ITensorAccessorUPtr accessor = nullptr); + static NodeID + add_output_node(Graph &g, NodeParams params, NodeIdxPair input, ITensorAccessorUPtr accessor = nullptr); /** Adds an activation layer node to the graph * * @param[in] g Graph to add the node to @@ -82,7 +85,10 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_activation_node(Graph &g, NodeParams params, NodeIdxPair input, ActivationLayerInfo act_info, + static NodeID add_activation_node(Graph &g, + NodeParams params, + NodeIdxPair input, + ActivationLayerInfo act_info, const QuantizationInfo &out_quant_info = QuantizationInfo()); /** Adds an activation layer node to the graph * @@ -96,7 +102,11 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_arg_min_max_node(Graph &g, NodeParams params, NodeIdxPair input, ReductionOperation op, unsigned int axis, + static NodeID add_arg_min_max_node(Graph &g, + NodeParams params, + NodeIdxPair input, + ReductionOperation op, + unsigned int axis, DataType out_data_type = DataType::UNKNOWN, const QuantizationInfo &out_quant_info = QuantizationInfo()); /** Adds a batch normalization layer node to the graph @@ -112,9 +122,14 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_batch_normalization_node(Graph &g, NodeParams params, NodeIdxPair input, float epsilon, - ITensorAccessorUPtr mean_accessor = nullptr, ITensorAccessorUPtr var_accessor = nullptr, - ITensorAccessorUPtr beta_accessor = nullptr, ITensorAccessorUPtr gamma_accessor = nullptr); + static NodeID add_batch_normalization_node(Graph &g, + NodeParams params, + NodeIdxPair input, + float epsilon, + ITensorAccessorUPtr mean_accessor = nullptr, + ITensorAccessorUPtr var_accessor = nullptr, + ITensorAccessorUPtr beta_accessor = nullptr, + ITensorAccessorUPtr gamma_accessor = nullptr); /** Adds a bounding box transform layer node to the graph * * @param[in] g Graph to add the node to @@ -125,7 +140,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_bounding_box_transform_node(Graph &g, NodeParams params, NodeIdxPair input, NodeIdxPair deltas, BoundingBoxTransformInfo info); + static NodeID add_bounding_box_transform_node( + Graph &g, NodeParams params, NodeIdxPair input, NodeIdxPair deltas, BoundingBoxTransformInfo info); /** Adds an channel shuffle layer node to the graph * * @param[in] g Graph to add the node to @@ -154,10 +170,17 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_convolution_node(Graph &g, NodeParams params, NodeIdxPair input, - Size2D kernel_spatial_extend, unsigned int depth, PadStrideInfo conv_info, unsigned int num_groups = 1, - ConvolutionMethod method = ConvolutionMethod::Default, FastMathHint fast_math_hint = FastMathHint::Disabled, - ITensorAccessorUPtr weights_accessor = nullptr, ITensorAccessorUPtr bias_accessor = nullptr, + static NodeID add_convolution_node(Graph &g, + NodeParams params, + NodeIdxPair input, + Size2D kernel_spatial_extend, + unsigned int depth, + PadStrideInfo conv_info, + unsigned int num_groups = 1, + ConvolutionMethod method = ConvolutionMethod::Default, + FastMathHint fast_math_hint = FastMathHint::Disabled, + ITensorAccessorUPtr weights_accessor = nullptr, + ITensorAccessorUPtr bias_accessor = nullptr, const QuantizationInfo &weights_quant_info = QuantizationInfo(), const QuantizationInfo &out_quant_info = QuantizationInfo()); /** Adds a deconvolution layer node to the graph @@ -173,9 +196,14 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_deconvolution_node(Graph &g, NodeParams params, NodeIdxPair input, - Size2D kernel_spatial_extend, unsigned int depth, PadStrideInfo deconv_info, - ITensorAccessorUPtr weights_accessor = nullptr, ITensorAccessorUPtr bias_accessor = nullptr); + static NodeID add_deconvolution_node(Graph &g, + NodeParams params, + NodeIdxPair input, + Size2D kernel_spatial_extend, + unsigned int depth, + PadStrideInfo deconv_info, + ITensorAccessorUPtr weights_accessor = nullptr, + ITensorAccessorUPtr bias_accessor = nullptr); /** Adds a depth concatenate node to the graph * * @param[in] g Graph to add the node to @@ -185,7 +213,10 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_concatenate_node(Graph &g, NodeParams params, const std::vector<NodeIdxPair> &inputs, const descriptors::ConcatLayerDescriptor &concat_descriptor); + static NodeID add_concatenate_node(Graph &g, + NodeParams params, + const std::vector<NodeIdxPair> &inputs, + const descriptors::ConcatLayerDescriptor &concat_descriptor); /** Adds an depth to space layer node to the graph * * @param[in] g Graph to add the node to @@ -212,11 +243,18 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_depthwise_convolution_node(Graph &g, NodeParams params, NodeIdxPair input, - Size2D kernel_spatial_extend, PadStrideInfo conv_info, int depth_multiplier = 1, - DepthwiseConvolutionMethod method = DepthwiseConvolutionMethod::Default, - ITensorAccessorUPtr weights_accessor = nullptr, ITensorAccessorUPtr bias_accessor = nullptr, const QuantizationInfo &quant_info = QuantizationInfo(), - const QuantizationInfo &out_quant_info = QuantizationInfo()); + static NodeID + add_depthwise_convolution_node(Graph &g, + NodeParams params, + NodeIdxPair input, + Size2D kernel_spatial_extend, + PadStrideInfo conv_info, + int depth_multiplier = 1, + DepthwiseConvolutionMethod method = DepthwiseConvolutionMethod::Default, + ITensorAccessorUPtr weights_accessor = nullptr, + ITensorAccessorUPtr bias_accessor = nullptr, + const QuantizationInfo &quant_info = QuantizationInfo(), + const QuantizationInfo &out_quant_info = QuantizationInfo()); /** Adds an element-wise layer node to the graph * * @param[in] g Graph to add the node to @@ -227,7 +265,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_elementwise_node(Graph &g, NodeParams params, NodeIdxPair input0, NodeIdxPair input1, EltwiseOperation operation); + static NodeID add_elementwise_node( + Graph &g, NodeParams params, NodeIdxPair input0, NodeIdxPair input1, EltwiseOperation operation); /** Adds a dequantization node to the graph * * @param[in] g Graph to add the node to @@ -248,7 +287,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_detection_output_node(Graph &g, NodeParams params, NodeIdxPair input_loc, NodeIdxPair input_conf, NodeIdxPair input_priorbox, const DetectionOutputLayerInfo &detect_info); + static NodeID add_detection_output_node(Graph &g, + NodeParams params, + NodeIdxPair input_loc, + NodeIdxPair input_conf, + NodeIdxPair input_priorbox, + const DetectionOutputLayerInfo &detect_info); /** Adds a detection post process layer node to the graph * * @param[in] g Graph to add the node to @@ -261,8 +305,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_detection_post_process_node(Graph &g, NodeParams params, NodeIdxPair input_box_encoding, NodeIdxPair input_class_prediction, - const DetectionPostProcessLayerInfo &detect_info, ITensorAccessorUPtr anchors_accessor = nullptr, + static NodeID add_detection_post_process_node(Graph &g, + NodeParams params, + NodeIdxPair input_box_encoding, + NodeIdxPair input_class_prediction, + const DetectionPostProcessLayerInfo &detect_info, + ITensorAccessorUPtr anchors_accessor = nullptr, const QuantizationInfo &anchor_quant_info = QuantizationInfo()); /** Adds a Dummy node to the graph * @@ -299,8 +347,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_fully_connected_layer(Graph &g, NodeParams params, NodeIdxPair input, unsigned int num_outputs, - NodeID weights_nid, NodeID bias_nid = EmptyNodeID, + static NodeID add_fully_connected_layer(Graph &g, + NodeParams params, + NodeIdxPair input, + unsigned int num_outputs, + NodeID weights_nid, + NodeID bias_nid = EmptyNodeID, const FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), const QuantizationInfo &out_quant_info = QuantizationInfo(), FastMathHint fast_math_hint = FastMathHint::Disabled); @@ -319,9 +371,13 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_fully_connected_layer(Graph &g, NodeParams params, NodeIdxPair input, unsigned int num_outputs, - ITensorAccessorUPtr weights_accessor = nullptr, ITensorAccessorUPtr bias_accessor = nullptr, - const FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), + static NodeID add_fully_connected_layer(Graph &g, + NodeParams params, + NodeIdxPair input, + unsigned int num_outputs, + ITensorAccessorUPtr weights_accessor = nullptr, + ITensorAccessorUPtr bias_accessor = nullptr, + const FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), const QuantizationInfo &weights_quant_info = QuantizationInfo(), const QuantizationInfo &out_quant_info = QuantizationInfo(), FastMathHint fast_math_hint = FastMathHint::Disabled); @@ -336,8 +392,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_generate_proposals_node(Graph &g, NodeParams params, NodeIdxPair scores, NodeIdxPair deltas, - NodeIdxPair anchors, GenerateProposalsInfo info); + static NodeID add_generate_proposals_node(Graph &g, + NodeParams params, + NodeIdxPair scores, + NodeIdxPair deltas, + NodeIdxPair anchors, + GenerateProposalsInfo info); /** Adds a L2 Normalize layer node to the graph * * @param[in] g Graph to add the node to @@ -358,7 +418,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_normalization_node(Graph &g, NodeParams params, NodeIdxPair input, NormalizationLayerInfo norm_info); + static NodeID + add_normalization_node(Graph &g, NodeParams params, NodeIdxPair input, NormalizationLayerInfo norm_info); /** Adds a normalize planar YUV layer node to the graph * * @param[in] g Graph to add the node to @@ -369,8 +430,11 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_normalize_planar_yuv_node(Graph &g, NodeParams params, NodeIdxPair input, - ITensorAccessorUPtr mean_accessor = nullptr, ITensorAccessorUPtr std_accessor = nullptr); + static NodeID add_normalize_planar_yuv_node(Graph &g, + NodeParams params, + NodeIdxPair input, + ITensorAccessorUPtr mean_accessor = nullptr, + ITensorAccessorUPtr std_accessor = nullptr); /** Adds a pad layer node to the graph * * @param[in] g Graph to add the node to @@ -382,7 +446,11 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_pad_node(Graph &g, NodeParams params, NodeIdxPair input, const PaddingList &paddings, PixelValue pad_value = PixelValue()); + static NodeID add_pad_node(Graph &g, + NodeParams params, + NodeIdxPair input, + const PaddingList &paddings, + PixelValue pad_value = PixelValue()); /** Adds a permute layer node to the graph * * @param[in] g Graph to add the node to @@ -394,7 +462,11 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_permute_node(Graph &g, NodeParams params, NodeIdxPair input, PermutationVector perm, DataLayout layout = DataLayout::UNKNOWN); + static NodeID add_permute_node(Graph &g, + NodeParams params, + NodeIdxPair input, + PermutationVector perm, + DataLayout layout = DataLayout::UNKNOWN); /** Adds a pooling layer node to the graph * * @param[in] g Graph to add the node to @@ -426,8 +498,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_print_node(Graph &g, NodeParams params, NodeIdxPair input, std::ostream &stream, const IOFormatInfo &format_info = IOFormatInfo(), - const std::function<ITensor *(ITensor *)> transform = nullptr); + static NodeID add_print_node(Graph &g, + NodeParams params, + NodeIdxPair input, + std::ostream &stream, + const IOFormatInfo &format_info = IOFormatInfo(), + const std::function<ITensor *(ITensor *)> transform = nullptr); /** Adds a priorbox layer node to the graph * * @param[in] g Graph to add the node to @@ -438,7 +514,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_priorbox_node(Graph &g, NodeParams params, NodeIdxPair input0, NodeIdxPair input1, const PriorBoxLayerInfo &prior_info); + static NodeID add_priorbox_node( + Graph &g, NodeParams params, NodeIdxPair input0, NodeIdxPair input1, const PriorBoxLayerInfo &prior_info); /** Adds a quantization layer node to the graph * * @param[in] g Graph to add the node to @@ -448,7 +525,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_quantization_node(Graph &g, NodeParams params, NodeIdxPair input, const QuantizationInfo &out_quant_info); + static NodeID + add_quantization_node(Graph &g, NodeParams params, NodeIdxPair input, const QuantizationInfo &out_quant_info); /** Adds a reduction sum layer node to the graph * * @param[in] g Graph to add the node to @@ -460,7 +538,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_reduction_operation_node(Graph &g, NodeParams params, NodeIdxPair input, ReductionOperation op, int axis, bool keep_dims = true); + static NodeID add_reduction_operation_node( + Graph &g, NodeParams params, NodeIdxPair input, ReductionOperation op, int axis, bool keep_dims = true); /** Adds a reorg layer node to the graph * * @param[in] g Graph to add the node to @@ -492,7 +571,12 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_resize_node(Graph &g, NodeParams params, NodeIdxPair input, InterpolationPolicy policy, float width_scale, float height_scale); + static NodeID add_resize_node(Graph &g, + NodeParams params, + NodeIdxPair input, + InterpolationPolicy policy, + float width_scale, + float height_scale); /** Adds a ROI align layer node to the graph * * @param[in] g Graph to add the node to @@ -503,7 +587,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_roi_align_node(Graph &g, NodeParams params, NodeIdxPair input, NodeIdxPair rois, ROIPoolingLayerInfo pool_info); + static NodeID + add_roi_align_node(Graph &g, NodeParams params, NodeIdxPair input, NodeIdxPair rois, ROIPoolingLayerInfo pool_info); /** Adds a scale layer node to the graph * This layer computes a product of the input with a scale (read from mul_accessor) and it applies an offset (read from add_accessor). * output = input * mul_w + add_w @@ -516,8 +601,11 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_scale_layer(Graph &g, const NodeParams ¶ms, NodeIdxPair input, - ITensorAccessorUPtr mul_accessor = nullptr, ITensorAccessorUPtr add_accessor = nullptr); + static NodeID add_scale_layer(Graph &g, + const NodeParams ¶ms, + NodeIdxPair input, + ITensorAccessorUPtr mul_accessor = nullptr, + ITensorAccessorUPtr add_accessor = nullptr); /** Adds a softmax node to the graph * * @param[in] g Graph to add the node to @@ -538,7 +626,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_slice_node(Graph &g, NodeParams params, NodeIdxPair input, Coordinates &starts, Coordinates &ends); + static NodeID + add_slice_node(Graph &g, NodeParams params, NodeIdxPair input, Coordinates &starts, Coordinates &ends); /** Adds a split node to the graph * * @param[in] g Graph to add the node to @@ -549,7 +638,8 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_split_node(Graph &g, NodeParams params, NodeIdxPair input, unsigned int num_splits, unsigned int axis = 0); + static NodeID + add_split_node(Graph &g, NodeParams params, NodeIdxPair input, unsigned int num_splits, unsigned int axis = 0); /** Adds a stack layer node to the graph * * @param[in] g Graph to add the node to @@ -572,7 +662,13 @@ public: * * @return Node ID of the created node, EmptyNodeID in case of error */ - static NodeID add_strided_slice_node(Graph &g, NodeParams params, NodeIdxPair input, Coordinates &starts, Coordinates &ends, BiStrides &strides, StridedSliceLayerInfo info); + static NodeID add_strided_slice_node(Graph &g, + NodeParams params, + NodeIdxPair input, + Coordinates &starts, + Coordinates &ends, + BiStrides &strides, + StridedSliceLayerInfo info); /** Adds a yolo layer to the graph * * @param[in] g Graph to add the node to diff --git a/arm_compute/graph/GraphContext.h b/arm_compute/graph/GraphContext.h index 7beb598646..68fbaf5478 100644 --- a/arm_compute/graph/GraphContext.h +++ b/arm_compute/graph/GraphContext.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_GRAPH_CONTEXT_H #include "arm_compute/graph/Types.h" - #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/IWeightsManager.h" @@ -39,18 +38,18 @@ namespace graph /** Contains structs required for memory management */ struct MemoryManagerContext { - Target target = { Target::UNSPECIFIED }; /**< Target */ - std::shared_ptr<arm_compute::IMemoryManager> intra_mm = { nullptr }; /**< Intra-function memory manager */ - std::shared_ptr<arm_compute::IMemoryManager> cross_mm = { nullptr }; /**< Cross-function memory manager */ - std::shared_ptr<arm_compute::IMemoryGroup> cross_group = { nullptr }; /**< Cross-function memory group */ - IAllocator *allocator = { nullptr }; /**< Backend allocator to use */ + Target target = {Target::UNSPECIFIED}; /**< Target */ + std::shared_ptr<arm_compute::IMemoryManager> intra_mm = {nullptr}; /**< Intra-function memory manager */ + std::shared_ptr<arm_compute::IMemoryManager> cross_mm = {nullptr}; /**< Cross-function memory manager */ + std::shared_ptr<arm_compute::IMemoryGroup> cross_group = {nullptr}; /**< Cross-function memory group */ + IAllocator *allocator = {nullptr}; /**< Backend allocator to use */ }; /** Contains structs required for weights management */ struct WeightsManagerContext { - Target target = { Target::UNSPECIFIED }; /**< Target */ - std::shared_ptr<arm_compute::IWeightsManager> wm = { nullptr }; /**< Weights manager */ + Target target = {Target::UNSPECIFIED}; /**< Target */ + std::shared_ptr<arm_compute::IWeightsManager> wm = {nullptr}; /**< Weights manager */ }; /** Graph context **/ @@ -125,7 +124,7 @@ public: void finalize(); private: - GraphConfig _config; /**< Graph configuration */ + GraphConfig _config; /**< Graph configuration */ std::map<Target, MemoryManagerContext> _memory_managers; /**< Memory managers for each target */ std::map<Target, WeightsManagerContext> _weights_managers; /**< Weights managers for each target */ }; diff --git a/arm_compute/graph/IDeviceBackend.h b/arm_compute/graph/IDeviceBackend.h index f84aac0ae0..8ae92e3177 100644 --- a/arm_compute/graph/IDeviceBackend.h +++ b/arm_compute/graph/IDeviceBackend.h @@ -88,7 +88,8 @@ public: * * @return Backend sub-tensor handle */ - virtual std::unique_ptr<ITensorHandle> create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) = 0; + virtual std::unique_ptr<ITensorHandle> + create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) = 0; /** Configure a backend Node * * @note This creates an appropriate configured backend function for the given node diff --git a/arm_compute/graph/LayerDescriptors.h b/arm_compute/graph/LayerDescriptors.h index c11174f2ce..d632ed9e78 100644 --- a/arm_compute/graph/LayerDescriptors.h +++ b/arm_compute/graph/LayerDescriptors.h @@ -37,8 +37,7 @@ namespace descriptors struct ConcatLayerDescriptor { /** Default constructor */ - ConcatLayerDescriptor() - : axis(DataLayoutDimension::CHANNEL), output_qinfo() + ConcatLayerDescriptor() : axis(DataLayoutDimension::CHANNEL), output_qinfo() { } @@ -46,8 +45,7 @@ struct ConcatLayerDescriptor * * @param[in] axis Axis. */ - ConcatLayerDescriptor(DataLayoutDimension axis) - : axis(axis), output_qinfo() + ConcatLayerDescriptor(DataLayoutDimension axis) : axis(axis), output_qinfo() { } @@ -76,9 +74,16 @@ struct EltwiseLayerDescriptor * @param[in] r_policy (Optional) Rounding policy used for the operation. Defaults to @ref RoundingPolicy::TO_ZERO * @param[in] fused_activation (Optional) Fused activation information. Defaults to empty (identity) @ref ActivationLayerInfo */ - EltwiseLayerDescriptor(EltwiseOperation op, QuantizationInfo out_quant_info = QuantizationInfo(), ConvertPolicy c_policy = ConvertPolicy::SATURATE, RoundingPolicy r_policy = RoundingPolicy::TO_ZERO, + EltwiseLayerDescriptor(EltwiseOperation op, + QuantizationInfo out_quant_info = QuantizationInfo(), + ConvertPolicy c_policy = ConvertPolicy::SATURATE, + RoundingPolicy r_policy = RoundingPolicy::TO_ZERO, ActivationLayerInfo fused_activation = ActivationLayerInfo()) - : op(op), out_quant_info(out_quant_info), c_policy(c_policy), r_policy(r_policy), fused_activation(fused_activation) + : op(op), + out_quant_info(out_quant_info), + c_policy(c_policy), + r_policy(r_policy), + fused_activation(fused_activation) { } @@ -100,10 +105,16 @@ struct UnaryEltwiseLayerDescriptor * @param[in] r_policy (Optional) Rounding policy used for the operation. Defaults to @ref RoundingPolicy::TO_ZERO * @param[in] fused_activation (Optional) Fused activation information. Defaults to empty (identity) @ref ActivationLayerInfo */ - UnaryEltwiseLayerDescriptor(UnaryEltwiseOperation op, QuantizationInfo out_quant_info = QuantizationInfo(), ConvertPolicy c_policy = ConvertPolicy::SATURATE, - RoundingPolicy r_policy = RoundingPolicy::TO_ZERO, - ActivationLayerInfo fused_activation = ActivationLayerInfo()) - : op(op), out_quant_info(out_quant_info), c_policy(c_policy), r_policy(r_policy), fused_activation(fused_activation) + UnaryEltwiseLayerDescriptor(UnaryEltwiseOperation op, + QuantizationInfo out_quant_info = QuantizationInfo(), + ConvertPolicy c_policy = ConvertPolicy::SATURATE, + RoundingPolicy r_policy = RoundingPolicy::TO_ZERO, + ActivationLayerInfo fused_activation = ActivationLayerInfo()) + : op(op), + out_quant_info(out_quant_info), + c_policy(c_policy), + r_policy(r_policy), + fused_activation(fused_activation) { } @@ -130,7 +141,7 @@ struct DeconvolutionLayerDescriptor PadStrideInfo info; /**< Padding and stride information */ QuantizationInfo out_quant_info; /**< Output quantization information */ }; -} // namespace descriptor +} // namespace descriptors } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_LAYER_DESCRIPTORS_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_LAYER_DESCRIPTORS_H */ diff --git a/arm_compute/graph/Logger.h b/arm_compute/graph/Logger.h index 872c650a1a..e83d5f4ddc 100644 --- a/arm_compute/graph/Logger.h +++ b/arm_compute/graph/Logger.h @@ -31,14 +31,14 @@ * * @note It will eventually create all default loggers in don't exist */ -#define ARM_COMPUTE_CREATE_DEFAULT_GRAPH_LOGGER() \ - do \ - { \ - if(arm_compute::logging::LoggerRegistry::get().logger("GRAPH") == nullptr) \ - { \ - arm_compute::logging::LoggerRegistry::get().create_reserved_loggers(); \ - } \ - } while(false) +#define ARM_COMPUTE_CREATE_DEFAULT_GRAPH_LOGGER() \ + do \ + { \ + if (arm_compute::logging::LoggerRegistry::get().logger("GRAPH") == nullptr) \ + { \ + arm_compute::logging::LoggerRegistry::get().create_reserved_loggers(); \ + } \ + } while (false) #else /* ARM_COMPUTE_LOGGING_ENABLED */ #define ARM_COMPUTE_CREATE_DEFAULT_GRAPH_LOGGER() #endif /* ARM_COMPUTE_LOGGING_ENABLED */ diff --git a/arm_compute/graph/Tensor.h b/arm_compute/graph/Tensor.h index de96c998bd..0ffae28ecc 100644 --- a/arm_compute/graph/Tensor.h +++ b/arm_compute/graph/Tensor.h @@ -24,11 +24,10 @@ #ifndef ARM_COMPUTE_GRAPH_TENSOR_H #define ARM_COMPUTE_GRAPH_TENSOR_H -#include "arm_compute/graph/Types.h" - #include "arm_compute/graph/ITensorAccessor.h" #include "arm_compute/graph/ITensorHandle.h" #include "arm_compute/graph/TensorDescriptor.h" +#include "arm_compute/graph/Types.h" #include <memory> #include <set> diff --git a/arm_compute/graph/TensorDescriptor.h b/arm_compute/graph/TensorDescriptor.h index 5fa155efc8..46a6ab2c27 100644 --- a/arm_compute/graph/TensorDescriptor.h +++ b/arm_compute/graph/TensorDescriptor.h @@ -52,7 +52,11 @@ struct TensorDescriptor final : public misc::ICloneable<TensorDescriptor> QuantizationInfo tensor_quant_info = QuantizationInfo(), DataLayout tensor_data_layout = DataLayout::NCHW, Target tensor_target = Target::UNSPECIFIED) - : shape(tensor_shape), data_type(tensor_data_type), layout(tensor_data_layout), quant_info(tensor_quant_info), target(tensor_target) + : shape(tensor_shape), + data_type(tensor_data_type), + layout(tensor_data_layout), + quant_info(tensor_quant_info), + target(tensor_target) { } /** Sets tensor descriptor shape @@ -106,11 +110,11 @@ struct TensorDescriptor final : public misc::ICloneable<TensorDescriptor> return std::make_unique<TensorDescriptor>(*this); } - TensorShape shape{}; /**< Tensor shape */ - DataType data_type{ DataType::UNKNOWN }; /**< Data type */ - DataLayout layout{ DataLayout::NCHW }; /**< Data layout */ - QuantizationInfo quant_info{}; /**< Quantization info */ - Target target{ Target::UNSPECIFIED }; /**< Target */ + TensorShape shape{}; /**< Tensor shape */ + DataType data_type{DataType::UNKNOWN}; /**< Data type */ + DataLayout layout{DataLayout::NCHW}; /**< Data layout */ + QuantizationInfo quant_info{}; /**< Quantization info */ + Target target{Target::UNSPECIFIED}; /**< Target */ }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/TypePrinter.h b/arm_compute/graph/TypePrinter.h index 9df4eba5ec..5e83820ab3 100644 --- a/arm_compute/graph/TypePrinter.h +++ b/arm_compute/graph/TypePrinter.h @@ -37,7 +37,7 @@ namespace graph /** Formatted output of the Target. */ inline ::std::ostream &operator<<(::std::ostream &os, const Target &target) { - switch(target) + switch (target) { case Target::UNSPECIFIED: os << "UNSPECIFIED"; @@ -60,7 +60,7 @@ inline ::std::ostream &operator<<(::std::ostream &os, const Target &target) inline ::std::ostream &operator<<(::std::ostream &os, const NodeType &node_type) { - switch(node_type) + switch (node_type) { case NodeType::ActivationLayer: os << "ActivationLayer"; @@ -207,7 +207,7 @@ inline ::std::ostream &operator<<(::std::ostream &os, const NodeType &node_type) /** Formatted output of the EltwiseOperation type. */ inline ::std::ostream &operator<<(::std::ostream &os, const EltwiseOperation &eltwise_op) { - switch(eltwise_op) + switch (eltwise_op) { case EltwiseOperation::Add: os << "Add"; @@ -231,7 +231,7 @@ inline ::std::ostream &operator<<(::std::ostream &os, const EltwiseOperation &el /** Formatted output of the ConvolutionMethod type. */ inline ::std::ostream &operator<<(::std::ostream &os, const ConvolutionMethod &method) { - switch(method) + switch (method) { case ConvolutionMethod::Default: os << "Default"; @@ -255,7 +255,7 @@ inline ::std::ostream &operator<<(::std::ostream &os, const ConvolutionMethod &m /** Formatted output of the FastMathHint type. */ inline ::std::ostream &operator<<(::std::ostream &os, const FastMathHint &hint) { - switch(hint) + switch (hint) { case FastMathHint::Enabled: os << "Enabled"; @@ -273,7 +273,7 @@ inline ::std::ostream &operator<<(::std::ostream &os, const FastMathHint &hint) /** Formatted output of the DepthwiseConvolutionMethod type. */ inline ::std::ostream &operator<<(::std::ostream &os, const DepthwiseConvolutionMethod &method) { - switch(method) + switch (method) { case DepthwiseConvolutionMethod::Default: os << "DEFAULT"; diff --git a/arm_compute/graph/Types.h b/arm_compute/graph/Types.h index 8d493403b3..5541e3cbcc 100644 --- a/arm_compute/graph/Types.h +++ b/arm_compute/graph/Types.h @@ -86,17 +86,18 @@ struct TensorDescriptor; /** Graph configuration structure */ struct GraphConfig { - bool use_function_memory_manager{ true }; /**< Use a memory manager to manage per-function auxilary memory */ - bool use_function_weights_manager{ true }; /**< Use a weights manager to manage transformed weights */ - bool use_transition_memory_manager{ true }; /**< Use a memory manager to manager transition buffer memory */ - bool use_tuner{ false }; /**< Use a tuner in tunable backends */ - bool use_synthetic_type{ false }; /**< Convert graph to a synthetic graph for a data type */ - DataType synthetic_type{ DataType::QASYMM8 }; /**< The data type of the synthetic graph */ - CLTunerMode tuner_mode{ CLTunerMode::EXHAUSTIVE }; /**< Tuner mode to be used by the CL tuner */ - int num_threads{ -1 }; /**< Number of threads to use (thread capable backends), if 0 the backend will auto-initialize, if -1 the backend will stay as it is. */ - std::string tuner_file{ "acl_tuner.csv" }; /**< File to load/store tuning values from */ - std::string mlgo_file{ "heuristics.mlgo" }; /**< Filename to load MLGO heuristics from */ - CLBackendType backend_type{ CLBackendType::Native }; /**< CL backend type to use */ + bool use_function_memory_manager{true}; /**< Use a memory manager to manage per-function auxilary memory */ + bool use_function_weights_manager{true}; /**< Use a weights manager to manage transformed weights */ + bool use_transition_memory_manager{true}; /**< Use a memory manager to manager transition buffer memory */ + bool use_tuner{false}; /**< Use a tuner in tunable backends */ + bool use_synthetic_type{false}; /**< Convert graph to a synthetic graph for a data type */ + DataType synthetic_type{DataType::QASYMM8}; /**< The data type of the synthetic graph */ + CLTunerMode tuner_mode{CLTunerMode::EXHAUSTIVE}; /**< Tuner mode to be used by the CL tuner */ + int num_threads{ + -1}; /**< Number of threads to use (thread capable backends), if 0 the backend will auto-initialize, if -1 the backend will stay as it is. */ + std::string tuner_file{"acl_tuner.csv"}; /**< File to load/store tuning values from */ + std::string mlgo_file{"heuristics.mlgo"}; /**< Filename to load MLGO heuristics from */ + CLBackendType backend_type{CLBackendType::Native}; /**< CL backend type to use */ }; /**< Device target types */ diff --git a/arm_compute/graph/Utils.h b/arm_compute/graph/Utils.h index a3d9012ee9..9813ff05c7 100644 --- a/arm_compute/graph/Utils.h +++ b/arm_compute/graph/Utils.h @@ -36,7 +36,7 @@ class GraphContext; inline bool is_utility_node(INode *node) { - std::set<NodeType> utility_node_types = { NodeType::PrintLayer }; + std::set<NodeType> utility_node_types = {NodeType::PrintLayer}; return utility_node_types.find(node->type()) != utility_node_types.end(); } diff --git a/arm_compute/graph/Workload.h b/arm_compute/graph/Workload.h index 5b4533cb6f..8ff0a548ae 100644 --- a/arm_compute/graph/Workload.h +++ b/arm_compute/graph/Workload.h @@ -69,8 +69,7 @@ public: */ struct ExecutionTask { - ExecutionTask(std::unique_ptr<arm_compute::IFunction> &&f, INode *n) - : task(std::move(f)), node(n) + ExecutionTask(std::unique_ptr<arm_compute::IFunction> &&f, INode *n) : task(std::move(f)), node(n) { } /** Prevent instances of this class from being copied (As this class contains pointers) */ @@ -97,11 +96,11 @@ struct ExecutionTask /** Execution workload */ struct ExecutionWorkload { - std::vector<Tensor *> inputs = {}; /**< Input handles */ - std::vector<Tensor *> outputs = {}; /**< Output handles */ - std::vector<ExecutionTask> tasks = {}; /**< Execution workload */ - Graph *graph = { nullptr }; /**< Graph bound to the workload */ - GraphContext *ctx = { nullptr }; /**< Graph execution context */ + std::vector<Tensor *> inputs = {}; /**< Input handles */ + std::vector<Tensor *> outputs = {}; /**< Output handles */ + std::vector<ExecutionTask> tasks = {}; /**< Execution workload */ + Graph *graph = {nullptr}; /**< Graph bound to the workload */ + GraphContext *ctx = {nullptr}; /**< Graph execution context */ }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/backends/BackendRegistrar.h b/arm_compute/graph/backends/BackendRegistrar.h index 902c12b0a6..2879361fef 100644 --- a/arm_compute/graph/backends/BackendRegistrar.h +++ b/arm_compute/graph/backends/BackendRegistrar.h @@ -24,8 +24,8 @@ #ifndef ARM_COMPUTE_GRAPH_BACKEND_REGISTRAR_H #define ARM_COMPUTE_GRAPH_BACKEND_REGISTRAR_H -#include "arm_compute/graph/Types.h" #include "arm_compute/graph/backends/BackendRegistry.h" +#include "arm_compute/graph/Types.h" #include <utility> @@ -58,4 +58,4 @@ inline BackendRegistrar<T>::BackendRegistrar(Target target) } // namespace backends } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_GRAPH_BACKEND_REGISTRAR_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_GRAPH_BACKEND_REGISTRAR_H */ diff --git a/arm_compute/graph/backends/CL/CLDeviceBackend.h b/arm_compute/graph/backends/CL/CLDeviceBackend.h index 63674ad794..09e19d7688 100644 --- a/arm_compute/graph/backends/CL/CLDeviceBackend.h +++ b/arm_compute/graph/backends/CL/CLDeviceBackend.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_CLDEVICEBACKEND_H #include "arm_compute/graph/IDeviceBackend.h" - #include "arm_compute/runtime/CL/CLBufferAllocator.h" #include "arm_compute/runtime/CL/CLGEMMHeuristicsHandle.h" #include "arm_compute/runtime/CL/CLTuner.h" @@ -59,22 +58,23 @@ public: void set_kernel_tuning_mode(CLTunerMode tuning_mode); // Inherited overridden methods - void initialize_backend() override; - void setup_backend_context(GraphContext &ctx) override; - void release_backend_context(GraphContext &ctx) override; + void initialize_backend() override; + void setup_backend_context(GraphContext &ctx) override; + void release_backend_context(GraphContext &ctx) override; bool is_backend_supported() override; IAllocator *backend_allocator() override; std::unique_ptr<ITensorHandle> create_tensor(const Tensor &tensor) override; - std::unique_ptr<ITensorHandle> create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) override; - std::unique_ptr<arm_compute::IFunction> configure_node(INode &node, GraphContext &ctx) override; - Status validate_node(INode &node) override; - std::shared_ptr<arm_compute::IMemoryManager> create_memory_manager(MemoryManagerAffinity affinity) override; + std::unique_ptr<ITensorHandle> + create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) override; + std::unique_ptr<arm_compute::IFunction> configure_node(INode &node, GraphContext &ctx) override; + Status validate_node(INode &node) override; + std::shared_ptr<arm_compute::IMemoryManager> create_memory_manager(MemoryManagerAffinity affinity) override; std::shared_ptr<arm_compute::IWeightsManager> create_weights_manager() override; void sync() override; private: - int _context_count; /**< Counts how many contexts are currently using the backend */ - CLTuner _tuner; /**< CL kernel tuner */ + int _context_count; /**< Counts how many contexts are currently using the backend */ + CLTuner _tuner; /**< CL kernel tuner */ CLGEMMHeuristicsHandle _gemm_heuristics; /**< GEMM heuristics */ std::unique_ptr<CLBufferAllocator> _allocator; /**< CL buffer affinity allocator */ std::string _tuner_file; /**< Filename to load/store the tuner's values from */ diff --git a/arm_compute/graph/backends/CL/CLSubTensorHandle.h b/arm_compute/graph/backends/CL/CLSubTensorHandle.h index 3750fc85ee..85eebec639 100644 --- a/arm_compute/graph/backends/CL/CLSubTensorHandle.h +++ b/arm_compute/graph/backends/CL/CLSubTensorHandle.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_CLSUBTENSORHANDLE_H #include "arm_compute/graph/ITensorHandle.h" - #include "arm_compute/runtime/CL/CLSubTensor.h" namespace arm_compute @@ -45,7 +44,10 @@ public: * @param[in] coords Starting coordinates * @param[in] extend_parent Extends parent shape if true */ - CLSubTensorHandle(ITensorHandle *parent_handle, const TensorShape &shape, const Coordinates &coords, bool extend_parent = false); + CLSubTensorHandle(ITensorHandle *parent_handle, + const TensorShape &shape, + const Coordinates &coords, + bool extend_parent = false); /** Destructor: free the tensor's memory */ ~CLSubTensorHandle() = default; /** Allow instances of this class to be move constructed */ @@ -58,10 +60,10 @@ public: CLSubTensorHandle &operator=(const CLSubTensorHandle &) = delete; // Inherited overridden methods - void allocate() override; - void free() override; - void manage(IMemoryGroup *mg) override; - void map(bool blocking) override; + void allocate() override; + void free() override; + void manage(IMemoryGroup *mg) override; + void map(bool blocking) override; void unmap() override; void release_if_unused() override; arm_compute::ITensor &tensor() override; diff --git a/arm_compute/graph/backends/CL/CLTensorHandle.h b/arm_compute/graph/backends/CL/CLTensorHandle.h index 16e30efc43..57e9794ec3 100644 --- a/arm_compute/graph/backends/CL/CLTensorHandle.h +++ b/arm_compute/graph/backends/CL/CLTensorHandle.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_CLTENSORHANDLE_H #include "arm_compute/graph/ITensorHandle.h" - #include "arm_compute/runtime/CL/CLTensor.h" namespace arm_compute @@ -51,10 +50,10 @@ public: CLTensorHandle &operator=(CLTensorHandle &&) = default; // Inherited overridden methods - void allocate() override; - void free() override; - void manage(IMemoryGroup *mg) override; - void map(bool blocking) override; + void allocate() override; + void free() override; + void manage(IMemoryGroup *mg) override; + void map(bool blocking) override; void unmap() override; void release_if_unused() override; arm_compute::ITensor &tensor() override; diff --git a/arm_compute/graph/backends/FunctionHelpers.h b/arm_compute/graph/backends/FunctionHelpers.h index 877e1f92e4..fd8b6b5a69 100644 --- a/arm_compute/graph/backends/FunctionHelpers.h +++ b/arm_compute/graph/backends/FunctionHelpers.h @@ -24,19 +24,19 @@ #ifndef ACL_ARM_COMPUTE_GRAPH_BACKENDS_FUNCTIONHELPERS_H #define ACL_ARM_COMPUTE_GRAPH_BACKENDS_FUNCTIONHELPERS_H +#include "arm_compute/core/Error.h" +#include "arm_compute/core/Helpers.h" +#include "arm_compute/core/ITensorInfo.h" +#include "arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h" +#include "arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h" +#include "arm_compute/graph/backends/Utils.h" #include "arm_compute/graph/Logger.h" +#include "arm_compute/graph/nodes/Nodes.h" #include "arm_compute/graph/Tensor.h" #include "arm_compute/graph/TypePrinter.h" #include "arm_compute/graph/Types.h" #include "arm_compute/graph/Utils.h" -#include "arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h" -#include "arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h" -#include "arm_compute/graph/backends/Utils.h" -#include "arm_compute/graph/nodes/Nodes.h" -#include "arm_compute/core/Error.h" -#include "arm_compute/core/Helpers.h" -#include "arm_compute/core/ITensorInfo.h" #include "support/Cast.h" namespace arm_compute @@ -59,13 +59,16 @@ template <typename TargetInfo> typename TargetInfo::TensorType *get_backing_tensor(arm_compute::graph::Tensor *tensor) { typename TargetInfo::TensorType *backing_tensor = nullptr; - if(tensor != nullptr) + if (tensor != nullptr) { ARM_COMPUTE_ERROR_ON(tensor->desc().target != TargetInfo::TargetType); // Get backing tensor handle ITensorHandle *tensor_handle = tensor->handle(); // Get backing tensor - backing_tensor = (tensor_handle != nullptr) ? arm_compute::utils::cast::polymorphic_cast<typename TargetInfo::TensorType *>(&tensor_handle->tensor()) : nullptr; + backing_tensor = (tensor_handle != nullptr) + ? arm_compute::utils::cast::polymorphic_cast<typename TargetInfo::TensorType *>( + &tensor_handle->tensor()) + : nullptr; } return backing_tensor; @@ -74,11 +77,8 @@ typename TargetInfo::TensorType *get_backing_tensor(arm_compute::graph::Tensor * template <typename TargetInfo> void validate_node(const INode &node, size_t num_expected_inputs, size_t num_expected_outputs) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating " << node.type() - << " Target: " << TargetInfo::TargetType - << " ID: " << node.id() - << node.name() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating " << node.type() << " Target: " << TargetInfo::TargetType + << " ID: " << node.id() << node.name() << std::endl); ARM_COMPUTE_ERROR_ON(TargetInfo::TargetType != node.assigned_target()); ARM_COMPUTE_ERROR_ON(node.num_inputs() != num_expected_inputs); @@ -109,17 +109,11 @@ std::unique_ptr<IFunction> create_activation_layer(ActivationLayerNode &node) auto func = std::make_unique<ActivationLayerFunction>(); func->configure(input, output, act_info); - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << " Activation function: " << act_info.activation() - << " a: " << act_info.a() - << " b: " << act_info.b() - << " InPlace : " << is_in_place_operation(input, output) - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO( + "Instantiated " << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Shape: " << input->info()->tensor_shape() + << " Activation function: " << act_info.activation() << " a: " << act_info.a() << " b: " + << act_info.b() << " InPlace : " << is_in_place_operation(input, output) << std::endl); return func; } @@ -148,15 +142,10 @@ std::unique_ptr<IFunction> create_arg_min_max_layer(ArgMinMaxLayerNode &node) auto func = std::make_unique<ArgMinMaxLayerFunction>(); func->configure(input, axis, output, op); - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << " Reduction Operation: " << op - << " axis: " << axis - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Shape: " << input->info()->tensor_shape() + << " Reduction Operation: " << op << " axis: " << axis << std::endl); return func; } @@ -191,16 +180,11 @@ std::unique_ptr<IFunction> create_batch_normalization_layer(BatchNormalizationLa func->configure(input, output, mean, var, beta, gamma, epsilon, fused_act); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << " Epsilon: " << epsilon << " " - << (fused_act.enabled() ? to_string(fused_act.activation()) : "") - << " InPlace: " << is_in_place_operation(input, output) - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Shape: " << input->info()->tensor_shape() << " Epsilon: " << epsilon + << " " << (fused_act.enabled() ? to_string(fused_act.activation()) : "") + << " InPlace: " << is_in_place_operation(input, output) << std::endl); return func; } @@ -216,7 +200,8 @@ std::unique_ptr<IFunction> create_batch_normalization_layer(BatchNormalizationLa * @return Backend batch normalization layer function */ template <typename FusedLayerTypes, typename TargetInfo> -std::unique_ptr<IFunction> create_fused_convolution_batch_normalization_layer(FusedConvolutionBatchNormalizationNode &node, GraphContext &ctx) +std::unique_ptr<IFunction> +create_fused_convolution_batch_normalization_layer(FusedConvolutionBatchNormalizationNode &node, GraphContext &ctx) { validate_node<TargetInfo>(node, 7 /* expected inputs */, 1 /* expected outputs */); @@ -246,19 +231,16 @@ std::unique_ptr<IFunction> create_fused_convolution_batch_normalization_layer(Fu // Create and configure function std::tie(func, func_name) = create_named_memory_managed_function<FType>( - std::string("FusedConvolutionBatchNormalizationLayer"), mm, input, weights, biases, output, mean, var, beta, gamma, epsilon, conv_info, num_groups, fast_math, fused_act); + std::string("FusedConvolutionBatchNormalizationLayer"), mm, input, weights, biases, output, mean, var, beta, + gamma, epsilon, conv_info, num_groups, fast_math, fused_act); // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Weights shape: " << weights->info()->tensor_shape() + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Input shape: " + << input->info()->tensor_shape() << " Weights shape: " << weights->info()->tensor_shape() << " Output shape: " << output->info()->tensor_shape() - << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") - << std::endl); + << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") << std::endl); return func; } @@ -273,7 +255,9 @@ std::unique_ptr<IFunction> create_fused_convolution_batch_normalization_layer(Fu * @return Backend fused depthwise convolution batch normalization layer function */ template <typename FusedLayerTypes, typename TargetInfo> -std::unique_ptr<IFunction> create_fused_depthwise_convolution_batch_normalization_layer(FusedDepthwiseConvolutionBatchNormalizationNode &node, GraphContext &ctx) +std::unique_ptr<IFunction> +create_fused_depthwise_convolution_batch_normalization_layer(FusedDepthwiseConvolutionBatchNormalizationNode &node, + GraphContext &ctx) { validate_node<TargetInfo>(node, 7 /* expected inputs */, 1 /* expected outputs */); @@ -302,19 +286,16 @@ std::unique_ptr<IFunction> create_fused_depthwise_convolution_batch_normalizatio // Create and configure function std::tie(func, func_name) = create_named_memory_managed_function<FType>( - std::string("FusedDepthwiseConvolutionBatchNormalizationLayer"), mm, input, weights, biases, output, mean, var, beta, gamma, epsilon, conv_info, depth_multiplier, fused_act); + std::string("FusedDepthwiseConvolutionBatchNormalizationLayer"), mm, input, weights, biases, output, mean, var, + beta, gamma, epsilon, conv_info, depth_multiplier, fused_act); // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Weights shape: " << weights->info()->tensor_shape() + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Input shape: " + << input->info()->tensor_shape() << " Weights shape: " << weights->info()->tensor_shape() << " Output shape: " << output->info()->tensor_shape() - << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") - << std::endl); + << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") << std::endl); return func; } @@ -343,15 +324,11 @@ std::unique_ptr<IFunction> create_bounding_box_transform_layer(BoundingBoxTransf func->configure(input, output, deltas, bbox_info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << " BoundingBox Info img W: " << bbox_info.img_width() << " " - << " BoundingBox Info img H: " << bbox_info.img_height() << " " - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO( + "Instantiated " << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Shape: " << input->info()->tensor_shape() + << " BoundingBox Info img W: " << bbox_info.img_width() << " " + << " BoundingBox Info img H: " << bbox_info.img_height() << " " << std::endl); return std::move(func); } @@ -379,14 +356,10 @@ std::unique_ptr<IFunction> create_channel_shuffle_layer(ChannelShuffleLayerNode auto func = std::make_unique<ChannelShuffleLayerFunction>(); func->configure(input, output, num_groups); - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << " Num groups: " << num_groups - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Shape: " << input->info()->tensor_shape() + << " Num groups: " << num_groups << std::endl); return func; } @@ -403,24 +376,25 @@ std::unique_ptr<IFunction> create_channel_shuffle_layer(ChannelShuffleLayerNode template <typename ConcatenateLayerFunction, typename TargetInfo> std::unique_ptr<arm_compute::IFunction> create_concatenate_layer(ConcatenateLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating Concatenate node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating Concatenate node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_ERROR_ON(node.num_outputs() != 1); // Return nullptr if depth concatenate is switched off - if(!node.is_enabled()) + if (!node.is_enabled()) { return nullptr; } // Extract IO and info std::vector<typename TargetInfo::SrcTensorType *> inputs; - for(unsigned int i = 0; i < node.num_inputs(); ++i) + for (unsigned int i = 0; i < node.num_inputs(); ++i) { inputs.push_back(get_backing_tensor<TargetInfo>(node.input(i))); } - typename TargetInfo::TensorType *output = get_backing_tensor<TargetInfo>(node.output(0)); - const DataLayout data_layout = node.output(0) != nullptr ? node.output(0)->desc().layout : DataLayout::UNKNOWN; - const size_t concat_axis = get_dimension_idx(data_layout, node.concatenation_axis()); + typename TargetInfo::TensorType *output = get_backing_tensor<TargetInfo>(node.output(0)); + const DataLayout data_layout = node.output(0) != nullptr ? node.output(0)->desc().layout : DataLayout::UNKNOWN; + const size_t concat_axis = get_dimension_idx(data_layout, node.concatenation_axis()); // Create and configure function auto func = std::make_unique<ConcatenateLayerFunction>(); @@ -429,20 +403,14 @@ std::unique_ptr<arm_compute::IFunction> create_concatenate_layer(ConcatenateLaye // Log info const bool is_quantized = is_data_type_quantized_asymmetric(output->info()->data_type()); std::ostringstream qss; - if(is_quantized) + if (is_quantized) { qss << " Output QuantInfo: " << output->info()->quantization_info(); } - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << output->info()->data_type() - << " Shape: " << output->info()->tensor_shape() - << " Num Inputs: " << inputs.size() - << " Axis: " << concat_axis - << qss.str() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO( + "Instantiated " << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << output->info()->data_type() << " Shape: " << output->info()->tensor_shape() + << " Num Inputs: " << inputs.size() << " Axis: " << concat_axis << qss.str() << std::endl); return func; } @@ -470,7 +438,7 @@ std::unique_ptr<IFunction> create_convolution_layer(ConvolutionLayerNode &node, const bool is_quantized = is_data_type_quantized_asymmetric(input->info()->data_type()); - if(is_quantized) + if (is_quantized) { biases->info()->set_data_type(DataType::S32); } @@ -486,55 +454,50 @@ std::unique_ptr<IFunction> create_convolution_layer(ConvolutionLayerNode &node, std::unique_ptr<IFunction> func; std::string func_name; - if(conv_algorithm == ConvolutionMethod::Winograd) + if (conv_algorithm == ConvolutionMethod::Winograd) { ARM_COMPUTE_ERROR_ON_MSG(num_groups != 1, "WinogradConvolutionLayer does not support grouping!"); - std::tie(func, func_name) = create_named_memory_managed_function<typename ConvolutionLayerFunctions::WinogradConvolutionLayer>( - std::string("WinogradConvolutionLayer"), mm, - input, weights, biases, output, conv_info, fused_act, fast_math); + std::tie(func, func_name) = + create_named_memory_managed_function<typename ConvolutionLayerFunctions::WinogradConvolutionLayer>( + std::string("WinogradConvolutionLayer"), mm, input, weights, biases, output, conv_info, fused_act, + fast_math); } - else if(conv_algorithm == ConvolutionMethod::Direct) + else if (conv_algorithm == ConvolutionMethod::Direct) { ARM_COMPUTE_ERROR_ON_MSG(num_groups != 1, "DirectConvolutionLayer does not support grouping!"); std::tie(func, func_name) = create_named_function<typename ConvolutionLayerFunctions::DirectConvolutionLayer>( - std::string("DirectConvolutionLayer"), - input, weights, biases, output, conv_info, fused_act); + std::string("DirectConvolutionLayer"), input, weights, biases, output, conv_info, fused_act); } - else if(conv_algorithm == ConvolutionMethod::GEMM) + else if (conv_algorithm == ConvolutionMethod::GEMM) { - std::tie(func, func_name) = create_named_memory_managed_function<typename ConvolutionLayerFunctions::GEMMConvolutionLayer>( - std::string("GEMMConvolutionLayer"), mm, - input, weights, biases, output, conv_info, - WeightsInfo(), Size2D(1U, 1U), fused_act, num_groups); + std::tie(func, func_name) = + create_named_memory_managed_function<typename ConvolutionLayerFunctions::GEMMConvolutionLayer>( + std::string("GEMMConvolutionLayer"), mm, input, weights, biases, output, conv_info, WeightsInfo(), + Size2D(1U, 1U), fused_act, num_groups); } else { - std::tie(func, func_name) = create_named_memory_managed_function<typename ConvolutionLayerFunctions::GenericConvolutionLayer>( - std::string("GenericConvolutionLayer"), mm, - input, weights, biases, output, conv_info, - WeightsInfo(), Size2D(1U, 1U), fused_act, fast_math, num_groups); + std::tie(func, func_name) = + create_named_memory_managed_function<typename ConvolutionLayerFunctions::GenericConvolutionLayer>( + std::string("GenericConvolutionLayer"), mm, input, weights, biases, output, conv_info, WeightsInfo(), + Size2D(1U, 1U), fused_act, fast_math, num_groups); } // Log info std::ostringstream qss; - if(is_quantized) + if (is_quantized) { qss << " Input QuantInfo: " << input->info()->quantization_info() << " Weights QuantInfo: " << weights->info()->quantization_info() << " Output QuantInfo: " << output->info()->quantization_info(); } ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << func_name - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Groups: " << num_groups + << node.name() << " Type: " << func_name << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Groups: " << num_groups << " Input shape: " << input->info()->tensor_shape() << " Weights shape: " << weights->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << qss.str() - << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") - << std::endl); + << " Output shape: " << output->info()->tensor_shape() << qss.str() + << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") << std::endl); return func; } @@ -566,19 +529,14 @@ std::unique_ptr<IFunction> create_deconvolution_layer(DeconvolutionLayerNode &no std::unique_ptr<IFunction> func; std::tie(func, std::ignore) = create_named_memory_managed_function<DeconvolutionLayerFunction>( - std::string(), mm, - input, weights, biases, output, deconv_info); + std::string(), mm, input, weights, biases, output, deconv_info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Weights shape: " << weights->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Weights shape: " << weights->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -604,7 +562,7 @@ std::unique_ptr<IFunction> create_depthwise_convolution_layer(DepthwiseConvoluti const bool is_quantized = is_data_type_quantized_asymmetric(input->info()->data_type()); - if(is_quantized) + if (is_quantized) { biases->info()->set_data_type(DataType::S32); } @@ -617,30 +575,25 @@ std::unique_ptr<IFunction> create_depthwise_convolution_layer(DepthwiseConvoluti std::unique_ptr<IFunction> func; std::string func_name; - std::tie(func, func_name) = create_named_function<DepthwiseConvolutionLayer>( - std::string("DepthwiseConvolutionLayer"), - input, weights, biases, output, conv_info, depth_multiplier, fused_act); + std::tie(func, func_name) = + create_named_function<DepthwiseConvolutionLayer>(std::string("DepthwiseConvolutionLayer"), input, weights, + biases, output, conv_info, depth_multiplier, fused_act); // Log info std::ostringstream qss; - if(is_quantized) + if (is_quantized) { qss << " Input QuantInfo: " << input->info()->quantization_info() << " Weights QuantInfo: " << weights->info()->quantization_info() << " Output QuantInfo: " << output->info()->quantization_info(); } ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << func_name - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Weights shape: " << weights->info()->tensor_shape() + << node.name() << " Type: " << func_name << " Target: " << TargetInfo::TargetType + << " Data Type: " << input->info()->data_type() << " Input shape: " + << input->info()->tensor_shape() << " Weights shape: " << weights->info()->tensor_shape() << " Output shape: " << output->info()->tensor_shape() - << " Depth multiplier: " << depth_multiplier - << qss.str() - << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") - << std::endl); + << " Depth multiplier: " << depth_multiplier << qss.str() + << (fused_act.enabled() ? " " + to_string(fused_act.activation()) : "") << std::endl); return func; } @@ -670,15 +623,11 @@ std::unique_ptr<IFunction> create_depth_to_space_layer(DepthToSpaceLayerNode &no func->configure(input, output, node.block_shape()); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Block Size: " << node.block_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Block Size: " << node.block_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -709,15 +658,11 @@ std::unique_ptr<IFunction> create_dequantization_layer(DequantizationLayerNode & func->configure(input, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Input quantization info: " << output->info()->quantization_info() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Input quantization info: " << output->info()->quantization_info() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -753,16 +698,12 @@ std::unique_ptr<IFunction> create_detection_output_layer(DetectionOutputLayerNod // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input0->info()->data_type() - << " Input0 shape: " << input0->info()->tensor_shape() - << " Input1 shape: " << input1->info()->tensor_shape() + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input0->info()->data_type() << " Input0 shape: " + << input0->info()->tensor_shape() << " Input1 shape: " << input1->info()->tensor_shape() << " Input2 shape: " << input2->info()->tensor_shape() << " Output shape: " << output->info()->tensor_shape() - << " DetectionOutputLayer info: " << detect_info - << std::endl); + << " DetectionOutputLayer info: " << detect_info << std::endl); return func; } @@ -805,19 +746,15 @@ std::unique_ptr<IFunction> create_detection_post_process_layer(DetectionPostProc // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input0->info()->data_type() - << " Input0 shape: " << input0->info()->tensor_shape() - << " Input1 shape: " << input1->info()->tensor_shape() + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input0->info()->data_type() << " Input0 shape: " + << input0->info()->tensor_shape() << " Input1 shape: " << input1->info()->tensor_shape() << " Input2 shape: " << input2->info()->tensor_shape() << " Output0 shape: " << output0->info()->tensor_shape() << " Output1 shape: " << output1->info()->tensor_shape() << " Output2 shape: " << output2->info()->tensor_shape() << " Output3 shape: " << output3->info()->tensor_shape() - << " DetectionPostProcessLayer info: " << detect_info - << std::endl); + << " DetectionPostProcessLayer info: " << detect_info << std::endl); return func; } @@ -849,35 +786,31 @@ std::unique_ptr<IFunction> create_eltwise_layer(EltwiseLayerNode &node) std::unique_ptr<IFunction> func = nullptr; std::string func_name; - if(eltwise_op == EltwiseOperation::Add) + if (eltwise_op == EltwiseOperation::Add) { std::tie(func, func_name) = create_named_function<typename EltwiseFunctions::Addition>( - std::string("ArithmeticAddition"), - input1, input2, output, convert_policy, act_info); + std::string("ArithmeticAddition"), input1, input2, output, convert_policy, act_info); } - else if(eltwise_op == EltwiseOperation::Sub) + else if (eltwise_op == EltwiseOperation::Sub) { std::tie(func, func_name) = create_named_function<typename EltwiseFunctions::Subtraction>( - std::string("ArithmeticSubtraction"), - input1, input2, output, convert_policy, act_info); + std::string("ArithmeticSubtraction"), input1, input2, output, convert_policy, act_info); } - else if(eltwise_op == EltwiseOperation::Mul) + else if (eltwise_op == EltwiseOperation::Mul) { std::tie(func, func_name) = create_named_function<typename EltwiseFunctions::Multiplication>( - std::string("PixelWiseMultiplication"), - input1, input2, output, 1.f, convert_policy, node.rounding_policy(), act_info); + std::string("PixelWiseMultiplication"), input1, input2, output, 1.f, convert_policy, node.rounding_policy(), + act_info); } - else if(eltwise_op == EltwiseOperation::Max) + else if (eltwise_op == EltwiseOperation::Max) { std::tie(func, func_name) = create_named_function<typename EltwiseFunctions::Maximum>( - std::string("ElementwiseMaximum"), - input1, input2, output, act_info); + std::string("ElementwiseMaximum"), input1, input2, output, act_info); } - else if(eltwise_op == EltwiseOperation::Div) + else if (eltwise_op == EltwiseOperation::Div) { std::tie(func, func_name) = create_named_function<typename EltwiseFunctions::Division>( - std::string("ArithmeticDivision"), - input1, input2, output, act_info); + std::string("ArithmeticDivision"), input1, input2, output, act_info); } else { @@ -885,14 +818,10 @@ std::unique_ptr<IFunction> create_eltwise_layer(EltwiseLayerNode &node) } // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Operation: " << func_name - << " Data Type: " << input1->info()->data_type() - << " Shape: " << input1->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() + << " Target: " << TargetInfo::TargetType << " Operation: " << func_name + << " Data Type: " << input1->info()->data_type() + << " Shape: " << input1->info()->tensor_shape() << std::endl); return func; } @@ -921,11 +850,10 @@ std::unique_ptr<IFunction> create_unary_eltwise_layer(UnaryEltwiseLayerNode &nod std::unique_ptr<IFunction> func = nullptr; std::string func_name; - if(eltwise_op == UnaryEltwiseOperation::Exp) + if (eltwise_op == UnaryEltwiseOperation::Exp) { - std::tie(func, func_name) = create_named_function<typename UnaryEltwiseFunctions::Exp>( - std::string("Exp"), - input, output); + std::tie(func, func_name) = + create_named_function<typename UnaryEltwiseFunctions::Exp>(std::string("Exp"), input, output); } else { @@ -933,14 +861,10 @@ std::unique_ptr<IFunction> create_unary_eltwise_layer(UnaryEltwiseLayerNode &nod } // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Operation: " << func_name - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() + << " Target: " << TargetInfo::TargetType << " Operation: " << func_name + << " Data Type: " << input->info()->data_type() + << " Shape: " << input->info()->tensor_shape() << std::endl); return func; } @@ -971,14 +895,10 @@ std::unique_ptr<IFunction> create_flatten_layer(FlattenLayerNode &node) func->configure(input, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1020,22 +940,17 @@ std::unique_ptr<IFunction> create_fully_connected_layer(FullyConnectedLayerNode // Log info std::ostringstream qss; - if(is_quantized) + if (is_quantized) { qss << " Input QuantInfo: " << input->info()->quantization_info() << " Weights QuantInfo: " << weights->info()->quantization_info() << " Output QuantInfo: " << output->info()->quantization_info(); } - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << qss.str() - << " Input shape: " << input->info()->tensor_shape() - << " Weights shape: " << weights->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << qss.str() << " Input shape: " << input->info()->tensor_shape() + << " Weights shape: " << weights->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1075,16 +990,14 @@ std::unique_ptr<IFunction> create_generate_proposals_layer(GenerateProposalsLaye func->configure(scores, deltas, anchors, proposals, scores_out, num_valid_proposals, info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.type() - << " Target " << TargetInfo::TargetType - << " Data Type: " << scores->info()->data_type() - << " Scores shape: " << scores->info()->tensor_shape() + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " + << node.type() << " Target " << TargetInfo::TargetType << " Data Type: " + << scores->info()->data_type() << " Scores shape: " << scores->info()->tensor_shape() << " Deltas shape: " << deltas->info()->tensor_shape() << " Anchors shape: " << anchors->info()->tensor_shape() << " Proposals shape: " << proposals->info()->tensor_shape() << " Num valid proposals shape: " << num_valid_proposals->info()->tensor_shape() - << " Scores Out shape: " << scores_out->info()->tensor_shape() - << std::endl); + << " Scores Out shape: " << scores_out->info()->tensor_shape() << std::endl); return std::move(func); } @@ -1119,16 +1032,11 @@ std::unique_ptr<IFunction> create_l2_normalize_layer(L2NormalizeLayerNode &node, func->configure(input, output, axis, epsilon); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Axis: " << axis - << " Epsilon: " << epsilon - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Axis: " << axis << " Epsilon: " << epsilon << std::endl); return func; } @@ -1162,15 +1070,11 @@ std::unique_ptr<IFunction> create_normalization_layer(NormalizationLayerNode &no func->configure(input, output, norm_info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Normalization info: " << norm_info.type() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Normalization info: " << norm_info.type() << std::endl); return std::move(func); } @@ -1204,13 +1108,9 @@ std::unique_ptr<IFunction> create_normalize_planar_yuv_layer(NormalizePlanarYUVL func->configure(input, output, mean, std); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Shape: " << input->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Shape: " << input->info()->tensor_shape() << std::endl); return std::move(func); } @@ -1242,14 +1142,10 @@ std::unique_ptr<IFunction> create_pad_layer(PadLayerNode &node) func->configure(input, output, padding, pad_value); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1280,15 +1176,11 @@ std::unique_ptr<IFunction> create_permute_layer(PermuteLayerNode &node) func->configure(input, output, perm); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Permutation vector: " << perm - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Permutation vector: " << perm << std::endl); return func; } @@ -1319,15 +1211,11 @@ std::unique_ptr<IFunction> create_pooling_layer(PoolingLayerNode &node) func->configure(input, output, pool_info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Pooling info: " << pool_info.pool_type - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Pooling info: " << pool_info.pool_type << std::endl); return func; } @@ -1358,14 +1246,10 @@ std::unique_ptr<IFunction> create_prelu_layer(PReluLayerNode &node) func->configure(input, alpha, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1388,13 +1272,9 @@ std::unique_ptr<IFunction> create_print_layer(PrintLayerNode &node) ARM_COMPUTE_UNUSED(input); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() << std::endl); return nullptr; } @@ -1428,15 +1308,11 @@ std::unique_ptr<IFunction> create_priorbox_layer(PriorBoxLayerNode &node) // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input0->info()->data_type() - << " Input0 shape: " << input0->info()->tensor_shape() - << " Input1 shape: " << input1->info()->tensor_shape() + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType + << " Data Type: " << input0->info()->data_type() << " Input0 shape: " + << input0->info()->tensor_shape() << " Input1 shape: " << input1->info()->tensor_shape() << " Output shape: " << output->info()->tensor_shape() - << " PriorBoxLayer info: " << prior_info - << std::endl); + << " PriorBoxLayer info: " << prior_info << std::endl); return func; } @@ -1466,14 +1342,10 @@ std::unique_ptr<IFunction> create_quantization_layer(QuantizationLayerNode &node func->configure(input, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1508,16 +1380,11 @@ std::unique_ptr<IFunction> create_reduction_operation_layer(ReductionLayerNode & // Log info ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType + << node.name() << " Type: " << node.type() << " Target: " << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Operation: " << op - << " Axis: " << axis - << " Keep dimensions:" << keep_dims - << std::endl); + << " Output shape: " << output->info()->tensor_shape() << " Operation: " << op + << " Axis: " << axis << " Keep dimensions:" << keep_dims << std::endl); return func; } @@ -1547,14 +1414,10 @@ std::unique_ptr<IFunction> create_reorg_layer(ReorgLayerNode &node) func->configure(input, output, node.stride()); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1584,14 +1447,10 @@ std::unique_ptr<IFunction> create_reshape_layer(ReshapeLayerNode &node) func->configure(input, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1619,18 +1478,15 @@ std::unique_ptr<IFunction> create_resize_layer(ResizeLayerNode &node) // Create and configure function auto func = std::make_unique<ResizeLayerFunction>(); - func->configure(input, output, ScaleKernelInfo{ policy, BorderMode::CONSTANT, PixelValue(), SamplingPolicy::CENTER, false, false }); + func->configure(input, output, + ScaleKernelInfo{policy, BorderMode::CONSTANT, PixelValue(), SamplingPolicy::CENTER, false, false}); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Interpolation: " << policy - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Interpolation: " << policy << std::endl); return func; } @@ -1665,17 +1521,13 @@ std::unique_ptr<IFunction> create_roi_align_layer(ROIAlignLayerNode &node) func->configure(input, rois, output, pool_info); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " ROIs shape: " << rois->info()->tensor_shape() - << " ROIPooling width: " << pool_info.pooled_width() - << " ROIPooling height: " << pool_info.pooled_height() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " ROIs shape: " << rois->info()->tensor_shape() + << " ROIPooling width: " << pool_info.pooled_width() + << " ROIPooling height: " << pool_info.pooled_height() << std::endl); return std::move(func); } @@ -1705,14 +1557,10 @@ std::unique_ptr<IFunction> create_slice_layer(SliceLayerNode &node) func->configure(input, output, node.starts(), node.ends()); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1744,14 +1592,10 @@ std::unique_ptr<IFunction> create_softmax_layer(SoftmaxLayerNode &node, GraphCon func->configure(input, output, beta); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } @@ -1768,12 +1612,13 @@ std::unique_ptr<IFunction> create_softmax_layer(SoftmaxLayerNode &node, GraphCon template <typename StackLayerFunction, typename TargetInfo> std::unique_ptr<arm_compute::IFunction> create_stack_layer(StackLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating Stack node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Creating Stack node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_ERROR_ON(node.num_outputs() != 1); // Extract IO and info std::vector<typename TargetInfo::TensorType *> inputs; - for(unsigned int i = 0; i < node.num_inputs(); ++i) + for (unsigned int i = 0; i < node.num_inputs(); ++i) { inputs.push_back(get_backing_tensor<TargetInfo>(node.input(i))); } @@ -1785,16 +1630,12 @@ std::unique_ptr<arm_compute::IFunction> create_stack_layer(StackLayerNode &node) func->configure(inputs, axis, output); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << output->info()->data_type() - << " Inputs shape: " << inputs[0]->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << " Num Inputs: " << inputs.size() - << " Axis: " << axis - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() + << " Target: " << TargetInfo::TargetType + << " Data Type: " << output->info()->data_type() + << " Inputs shape: " << inputs[0]->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() + << " Num Inputs: " << inputs.size() << " Axis: " << axis << std::endl); return func; } @@ -1829,14 +1670,10 @@ std::unique_ptr<IFunction> create_strided_slice_layer(StridedSliceLayerNode &nod func->configure(input, output, starts, ends, strides, info.begin_mask(), info.end_mask(), info.shrink_axis_mask()); // Log info - ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " - << node.name() - << " Type: " << node.type() - << " Target: " << TargetInfo::TargetType - << " Data Type: " << input->info()->data_type() - << " Input shape: " << input->info()->tensor_shape() - << " Output shape: " << output->info()->tensor_shape() - << std::endl); + ARM_COMPUTE_LOG_GRAPH_INFO("Instantiated " << node.name() << " Type: " << node.type() << " Target: " + << TargetInfo::TargetType << " Data Type: " << input->info()->data_type() + << " Input shape: " << input->info()->tensor_shape() + << " Output shape: " << output->info()->tensor_shape() << std::endl); return func; } diff --git a/arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h b/arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h index 19c627d479..27e21cbc7e 100644 --- a/arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h +++ b/arm_compute/graph/backends/FusedConvolutionBatchNormalizationFunction.h @@ -70,15 +70,19 @@ public: * @param[in] fused_act Activation layer information in case of a fused activation. * */ - void configure(TensorType *input, - TensorType *weights, - TensorType *bias, - TensorType *output, - const TensorType *mean, - const TensorType *var, - const TensorType *beta, - const TensorType *gamma, - float epsilon, const PadStrideInfo &conv_info, unsigned int num_groups, bool fast_math, ActivationLayerInfo const &fused_act) + void configure(TensorType *input, + TensorType *weights, + TensorType *bias, + TensorType *output, + const TensorType *mean, + const TensorType *var, + const TensorType *beta, + const TensorType *gamma, + float epsilon, + const PadStrideInfo &conv_info, + unsigned int num_groups, + bool fast_math, + ActivationLayerInfo const &fused_act) { // We don't run any validate, as we assume that the layers have been already validated const bool has_bias = (bias != nullptr); @@ -86,7 +90,7 @@ public: // We check if the layer has a bias. If yes, use it in-place. If not, we need to create one // as batch normalization might end up with a bias != 0 - if(has_bias) + if (has_bias) { _fused_batch_norm_layer.configure(weights, mean, var, nullptr, nullptr, bias, beta, gamma, epsilon); bias_to_use = bias; @@ -97,9 +101,10 @@ public: bias_to_use = &_fused_bias; } - _conv_layer.configure(input, weights, bias_to_use, output, conv_info, WeightsInfo(), Size2D(1U, 1U), fused_act, fast_math, num_groups); + _conv_layer.configure(input, weights, bias_to_use, output, conv_info, WeightsInfo(), Size2D(1U, 1U), fused_act, + fast_math, num_groups); - if(!has_bias) + if (!has_bias) { _fused_bias.allocator()->allocate(); } @@ -114,7 +119,7 @@ public: void prepare() { - if(!_is_prepared) + if (!_is_prepared) { _fused_batch_norm_layer.run(); _is_prepared = true; diff --git a/arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h b/arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h index 4f8a8da1fb..07a2cdd8b8 100644 --- a/arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h +++ b/arm_compute/graph/backends/FusedDepthwiseConvolutionBatchNormalizationFunction.h @@ -67,15 +67,18 @@ public: * @param[in] fused_act Activation layer information in case of a fused activation. * */ - void configure(TensorType *input, - TensorType *weights, - TensorType *bias, - TensorType *output, - const TensorType *mean, - const TensorType *var, - const TensorType *beta, - const TensorType *gamma, - float epsilon, const PadStrideInfo &conv_info, unsigned int depth_multiplier, ActivationLayerInfo const &fused_act) + void configure(TensorType *input, + TensorType *weights, + TensorType *bias, + TensorType *output, + const TensorType *mean, + const TensorType *var, + const TensorType *beta, + const TensorType *gamma, + float epsilon, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier, + ActivationLayerInfo const &fused_act) { // We don't run any validate, as we assume that the layers have been already validated const bool has_bias = (bias != nullptr); @@ -83,20 +86,23 @@ public: // We check if the layer has a bias. If yes, use it in-place. If not, we need to create one // as batch normalization might end up with a bias != 0 - if(has_bias) + if (has_bias) { - _fused_batch_norm_layer.configure(weights, mean, var, nullptr, nullptr, bias, beta, gamma, epsilon, FuseBatchNormalizationType::DEPTHWISECONVOLUTION); + _fused_batch_norm_layer.configure(weights, mean, var, nullptr, nullptr, bias, beta, gamma, epsilon, + FuseBatchNormalizationType::DEPTHWISECONVOLUTION); bias_to_use = bias; } else { - _fused_batch_norm_layer.configure(weights, mean, var, nullptr, &_fused_bias, nullptr, beta, gamma, epsilon, FuseBatchNormalizationType::DEPTHWISECONVOLUTION); + _fused_batch_norm_layer.configure(weights, mean, var, nullptr, &_fused_bias, nullptr, beta, gamma, epsilon, + FuseBatchNormalizationType::DEPTHWISECONVOLUTION); bias_to_use = &_fused_bias; } - _depth_conv_layer.configure(input, weights, bias_to_use, output, conv_info, depth_multiplier, fused_act.enabled() ? fused_act : ActivationLayerInfo()); + _depth_conv_layer.configure(input, weights, bias_to_use, output, conv_info, depth_multiplier, + fused_act.enabled() ? fused_act : ActivationLayerInfo()); - if(!has_bias) + if (!has_bias) { _fused_bias.allocator()->allocate(); } @@ -111,7 +117,7 @@ public: void prepare() { - if(!_is_prepared) + if (!_is_prepared) { _fused_batch_norm_layer.run(); _is_prepared = true; diff --git a/arm_compute/graph/backends/NEON/NEDeviceBackend.h b/arm_compute/graph/backends/NEON/NEDeviceBackend.h index 9cb37d4553..cd817a20d8 100644 --- a/arm_compute/graph/backends/NEON/NEDeviceBackend.h +++ b/arm_compute/graph/backends/NEON/NEDeviceBackend.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_NEDEVICEBACKEND_H #include "arm_compute/graph/IDeviceBackend.h" - #include "arm_compute/runtime/Allocator.h" namespace arm_compute @@ -41,16 +40,17 @@ public: NEDeviceBackend(); // Inherited overridden methods - void initialize_backend() override; - void setup_backend_context(GraphContext &ctx) override; - void release_backend_context(GraphContext &ctx) override; + void initialize_backend() override; + void setup_backend_context(GraphContext &ctx) override; + void release_backend_context(GraphContext &ctx) override; bool is_backend_supported() override; IAllocator *backend_allocator() override; std::unique_ptr<ITensorHandle> create_tensor(const Tensor &tensor) override; - std::unique_ptr<ITensorHandle> create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) override; - std::unique_ptr<arm_compute::IFunction> configure_node(INode &node, GraphContext &ctx) override; - Status validate_node(INode &node) override; - std::shared_ptr<arm_compute::IMemoryManager> create_memory_manager(MemoryManagerAffinity affinity) override; + std::unique_ptr<ITensorHandle> + create_subtensor(ITensorHandle *parent, TensorShape shape, Coordinates coords, bool extend_parent) override; + std::unique_ptr<arm_compute::IFunction> configure_node(INode &node, GraphContext &ctx) override; + Status validate_node(INode &node) override; + std::shared_ptr<arm_compute::IMemoryManager> create_memory_manager(MemoryManagerAffinity affinity) override; std::shared_ptr<arm_compute::IWeightsManager> create_weights_manager() override; void sync() override; diff --git a/arm_compute/graph/backends/NEON/NESubTensorHandle.h b/arm_compute/graph/backends/NEON/NESubTensorHandle.h index a438b65735..3619f4ed1b 100644 --- a/arm_compute/graph/backends/NEON/NESubTensorHandle.h +++ b/arm_compute/graph/backends/NEON/NESubTensorHandle.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_NESUBTENSORHANDLE_H #include "arm_compute/graph/ITensorHandle.h" - #include "arm_compute/runtime/SubTensor.h" namespace arm_compute @@ -45,7 +44,10 @@ public: * @param[in] coords Starting coordinates * @param[in] extend_parent Extends parent shape if true */ - NESubTensorHandle(ITensorHandle *parent_handle, const TensorShape &shape, const Coordinates &coords, bool extend_parent = false); + NESubTensorHandle(ITensorHandle *parent_handle, + const TensorShape &shape, + const Coordinates &coords, + bool extend_parent = false); /** Destructor: free the tensor's memory */ ~NESubTensorHandle() = default; /** Allow instances of this class to be move constructed */ @@ -58,10 +60,10 @@ public: NESubTensorHandle &operator=(const NESubTensorHandle &) = delete; // Inherited overridden methods - void allocate() override; - void free() override; - void manage(IMemoryGroup *mg) override; - void map(bool blocking) override; + void allocate() override; + void free() override; + void manage(IMemoryGroup *mg) override; + void map(bool blocking) override; void unmap() override; void release_if_unused() override; arm_compute::ITensor &tensor() override; diff --git a/arm_compute/graph/backends/NEON/NETensorHandle.h b/arm_compute/graph/backends/NEON/NETensorHandle.h index 99101a8fe9..1df90822ba 100644 --- a/arm_compute/graph/backends/NEON/NETensorHandle.h +++ b/arm_compute/graph/backends/NEON/NETensorHandle.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_GRAPH_NETENSORHANDLE_H #include "arm_compute/graph/ITensorHandle.h" - #include "arm_compute/runtime/Tensor.h" namespace arm_compute @@ -51,10 +50,10 @@ public: NETensorHandle &operator=(NETensorHandle &&) = default; // Inherited overridden methods - void allocate() override; - void free() override; - void manage(IMemoryGroup *mg) override; - void map(bool blocking) override; + void allocate() override; + void free() override; + void manage(IMemoryGroup *mg) override; + void map(bool blocking) override; void unmap() override; void release_if_unused() override; arm_compute::ITensor &tensor() override; diff --git a/arm_compute/graph/backends/Utils.h b/arm_compute/graph/backends/Utils.h index 774ce515b5..5f4e66c207 100644 --- a/arm_compute/graph/backends/Utils.h +++ b/arm_compute/graph/backends/Utils.h @@ -42,7 +42,8 @@ namespace backends * @return A configured backend function */ template <typename FunctionType, typename FunctionNameType, typename... ParameterType> -std::tuple<std::unique_ptr<arm_compute::IFunction>, FunctionNameType> create_named_function(FunctionNameType name, ParameterType... args) +std::tuple<std::unique_ptr<arm_compute::IFunction>, FunctionNameType> create_named_function(FunctionNameType name, + ParameterType... args) { auto f = std::make_unique<FunctionType>(); f->configure(std::forward<ParameterType>(args)...); @@ -58,9 +59,8 @@ std::tuple<std::unique_ptr<arm_compute::IFunction>, FunctionNameType> create_nam * @return A configured backend function */ template <typename FunctionType, typename FunctionNameType, typename MemoryManagerType, typename... ParameterType> -std::tuple<std::unique_ptr<arm_compute::IFunction>, FunctionNameType> create_named_memory_managed_function(FunctionNameType name, - MemoryManagerType mm, - ParameterType... args) +std::tuple<std::unique_ptr<arm_compute::IFunction>, FunctionNameType> +create_named_memory_managed_function(FunctionNameType name, MemoryManagerType mm, ParameterType... args) { auto f = std::make_unique<FunctionType>(mm); f->configure(std::forward<ParameterType>(args)...); diff --git a/arm_compute/graph/backends/ValidateHelpers.h b/arm_compute/graph/backends/ValidateHelpers.h index 71a6201554..0e102942a7 100644 --- a/arm_compute/graph/backends/ValidateHelpers.h +++ b/arm_compute/graph/backends/ValidateHelpers.h @@ -24,14 +24,13 @@ #ifndef ACL_ARM_COMPUTE_GRAPH_BACKENDS_VALIDATEHELPERS_H #define ACL_ARM_COMPUTE_GRAPH_BACKENDS_VALIDATEHELPERS_H -#include "arm_compute/graph/Logger.h" -#include "arm_compute/graph/Tensor.h" -#include "arm_compute/graph/Types.h" -#include "arm_compute/graph/nodes/Nodes.h" - #include "arm_compute/core/Error.h" #include "arm_compute/core/Helpers.h" #include "arm_compute/core/ITensorInfo.h" +#include "arm_compute/graph/Logger.h" +#include "arm_compute/graph/nodes/Nodes.h" +#include "arm_compute/graph/Tensor.h" +#include "arm_compute/graph/Types.h" namespace arm_compute { @@ -63,7 +62,8 @@ inline arm_compute::ITensorInfo *get_backing_tensor_info(arm_compute::graph::Ten template <typename ArgMinMaxLayer> Status validate_arg_min_max_layer(ArgMinMaxLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ArgMinMaxLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating ArgMinMaxLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -86,7 +86,8 @@ Status validate_arg_min_max_layer(ArgMinMaxLayerNode &node) template <typename BoundingBoxTransformLayer> Status validate_bounding_box_transform_layer(BoundingBoxTransformLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating BoundingBoxTransformLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating BoundingBoxTransformLayer node with ID : " << node.id() << " and Name: " + << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 2); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -110,7 +111,8 @@ Status validate_bounding_box_transform_layer(BoundingBoxTransformLayerNode &node template <typename ChannelShuffleLayer> Status validate_channel_shuffle_layer(ChannelShuffleLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ChannelShuffle node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating ChannelShuffle node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -133,10 +135,14 @@ Status validate_channel_shuffle_layer(ChannelShuffleLayerNode &node) * * @return Status */ -template <typename ConvolutionLayer, typename DirectConvolutionLayer, typename GEMMConvolutionLayer, typename WinogradConvolutionLayer> +template <typename ConvolutionLayer, + typename DirectConvolutionLayer, + typename GEMMConvolutionLayer, + typename WinogradConvolutionLayer> Status validate_convolution_layer(ConvolutionLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ConvolutionLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating ConvolutionLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -146,7 +152,7 @@ Status validate_convolution_layer(ConvolutionLayerNode &node) arm_compute::ITensorInfo *biases = get_backing_tensor_info(node.input(2)); arm_compute::ITensorInfo *output = get_backing_tensor_info(node.output(0)); - if(is_data_type_quantized_asymmetric(input->data_type())) + if (is_data_type_quantized_asymmetric(input->data_type())) { biases->set_data_type(DataType::S32); } @@ -158,23 +164,24 @@ Status validate_convolution_layer(ConvolutionLayerNode &node) // Validate function Status status{}; - switch(conv_algorithm) + switch (conv_algorithm) { case ConvolutionMethod::Direct: ARM_COMPUTE_RETURN_ERROR_ON_MSG(num_groups != 1, "DirectConvolutionLayer does not support grouping!"); status = DirectConvolutionLayer::validate(input, weights, biases, output, conv_info); break; case ConvolutionMethod::GEMM: - status = GEMMConvolutionLayer::validate(input, weights, biases, output, conv_info, - WeightsInfo(), Size2D(1, 1), ActivationLayerInfo(), num_groups); + status = GEMMConvolutionLayer::validate(input, weights, biases, output, conv_info, WeightsInfo(), + Size2D(1, 1), ActivationLayerInfo(), num_groups); break; case ConvolutionMethod::Winograd: ARM_COMPUTE_RETURN_ERROR_ON_MSG(num_groups != 1, "WinogradConvolutionLayer does not support grouping!"); - status = WinogradConvolutionLayer::validate(input, weights, biases, output, conv_info, ActivationLayerInfo(), fast_math); + status = WinogradConvolutionLayer::validate(input, weights, biases, output, conv_info, + ActivationLayerInfo(), fast_math); break; case ConvolutionMethod::Default: - status = ConvolutionLayer::validate(input, weights, biases, output, conv_info, - WeightsInfo(), Size2D(1, 1), ActivationLayerInfo(), fast_math, num_groups); + status = ConvolutionLayer::validate(input, weights, biases, output, conv_info, WeightsInfo(), Size2D(1, 1), + ActivationLayerInfo(), fast_math, num_groups); break; default: ARM_COMPUTE_RETURN_ERROR_MSG("Unsupported convolution method"); @@ -194,7 +201,8 @@ Status validate_convolution_layer(ConvolutionLayerNode &node) template <typename DepthwiseConvolutionLayer> Status validate_depthwise_convolution_layer(DepthwiseConvolutionLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DepthwiseConvolutionLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DepthwiseConvolutionLayer node with ID : " << node.id() << " and Name: " + << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -210,7 +218,7 @@ Status validate_depthwise_convolution_layer(DepthwiseConvolutionLayerNode &node) // Validate function Status status{}; - switch(dwc_algorithm) + switch (dwc_algorithm) { case DepthwiseConvolutionMethod::Default: case DepthwiseConvolutionMethod::Optimized3x3: @@ -233,7 +241,8 @@ Status validate_depthwise_convolution_layer(DepthwiseConvolutionLayerNode &node) template <typename DepthToSpaceLayer> Status validate_depth_to_space_layer(DepthToSpaceLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -254,7 +263,8 @@ Status validate_depth_to_space_layer(DepthToSpaceLayerNode &node) template <typename DequantizationLayer> Status validate_dequantization_layer(DequantizationLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -275,7 +285,8 @@ Status validate_dequantization_layer(DequantizationLayerNode &node) template <typename DetectionOutputLayer> Status validate_detection_output_layer(DetectionOutputLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating DetectionOutputLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -299,7 +310,8 @@ Status validate_detection_output_layer(DetectionOutputLayerNode &node) template <typename DetectionPostProcessLayer> Status validate_detection_post_process_layer(DetectionPostProcessLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DetectionPostProcessLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating DetectionPostProcessLayer node with ID : " << node.id() << " and Name: " + << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 4); @@ -327,7 +339,8 @@ Status validate_detection_post_process_layer(DetectionPostProcessLayerNode &node template <typename GenerateProposalsLayer> Status validate_generate_proposals_layer(GenerateProposalsLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating GenerateProposalsLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating GenerateProposalsLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 3); @@ -354,7 +367,8 @@ Status validate_generate_proposals_layer(GenerateProposalsLayerNode &node) template <typename L2NormalizeLayer> Status validate_l2_normalize_layer(L2NormalizeLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating L2NormalizeLayerNode node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating L2NormalizeLayerNode node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -379,7 +393,8 @@ Status validate_l2_normalize_layer(L2NormalizeLayerNode &node) template <typename NormalizePlanarYUVLayer> Status validate_normalize_planar_yuv_layer(NormalizePlanarYUVLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating NormalizePlanarYUVLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating NormalizePlanarYUVLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 3); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -404,7 +419,8 @@ Status validate_normalize_planar_yuv_layer(NormalizePlanarYUVLayerNode &node) template <typename PadLayer> Status validate_pad_layer(PadLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PadLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PadLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -427,14 +443,15 @@ Status validate_pad_layer(PadLayerNode &node) template <typename PermuteLayer> Status validate_permute_layer(PermuteLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PermuteLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PermuteLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); // Extract IO and info arm_compute::ITensorInfo *input = get_backing_tensor_info(node.input(0)); arm_compute::ITensorInfo *output = get_backing_tensor_info(node.output(0)); - const PermutationVector &perm = node.permutation_vector(); + const PermutationVector &perm = node.permutation_vector(); return PermuteLayer::validate(input, output, perm); } @@ -450,7 +467,8 @@ Status validate_permute_layer(PermuteLayerNode &node) template <typename PReluLayer> Status validate_prelu_layer(PReluLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PRelu node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PRelu node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 2); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -473,7 +491,8 @@ Status validate_prelu_layer(PReluLayerNode &node) template <typename PriorBoxLayer> Status validate_priorbox_layer(PriorBoxLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating PriorBoxLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating PriorBoxLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 2); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -497,7 +516,8 @@ Status validate_priorbox_layer(PriorBoxLayerNode &node) template <typename QuantizationLayer> Status validate_quantization_layer(QuantizationLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating QuantizationLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating QuantizationLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -520,7 +540,8 @@ Status validate_quantization_layer(QuantizationLayerNode &node) template <typename ReductionLayer> Status validate_reduction_operation_layer(ReductionLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ReductionLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating ReductionLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -544,7 +565,8 @@ Status validate_reduction_operation_layer(ReductionLayerNode &node) template <typename ReorgLayer> Status validate_reorg_layer(ReorgLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ReorgLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ReorgLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -567,7 +589,8 @@ Status validate_reorg_layer(ReorgLayerNode &node) template <typename ReshapeLayer> Status validate_reshape_layer(ReshapeLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ReshapeLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ReshapeLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -590,14 +613,15 @@ Status validate_reshape_layer(ReshapeLayerNode &node) template <typename ROIAlignLayer> Status validate_roi_align_layer(ROIAlignLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating ROIAlignLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE( + "Validating ROIAlignLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 2); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); // Extract input and output - arm_compute::ITensorInfo *input = detail::get_backing_tensor_info(node.input(0)); - arm_compute::ITensorInfo *rois = detail::get_backing_tensor_info(node.input(1)); - arm_compute::ITensorInfo *output = detail::get_backing_tensor_info(node.output(0)); + arm_compute::ITensorInfo *input = detail::get_backing_tensor_info(node.input(0)); + arm_compute::ITensorInfo *rois = detail::get_backing_tensor_info(node.input(1)); + arm_compute::ITensorInfo *output = detail::get_backing_tensor_info(node.output(0)); const ROIPoolingLayerInfo &pool_info = node.pooling_info(); // Validate function @@ -615,7 +639,8 @@ Status validate_roi_align_layer(ROIAlignLayerNode &node) template <typename SliceLayer> Status validate_slice_layer(SliceLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating Slice node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating Slice node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -639,7 +664,8 @@ Status validate_slice_layer(SliceLayerNode &node) template <typename StridedSliceLayer> Status validate_strided_slice_layer(StridedSliceLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating StridedSlice node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating StridedSlice node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -651,7 +677,8 @@ Status validate_strided_slice_layer(StridedSliceLayerNode &node) const BiStrides strides = node.strides(); const StridedSliceLayerInfo info = node.strided_slice_info(); - return StridedSliceLayer::validate(input, output, starts, ends, strides, info.begin_mask(), info.end_mask(), info.shrink_axis_mask()); + return StridedSliceLayer::validate(input, output, starts, ends, strides, info.begin_mask(), info.end_mask(), + info.shrink_axis_mask()); } /** Validates a element-wise layer node @@ -663,7 +690,8 @@ Status validate_strided_slice_layer(StridedSliceLayerNode &node) template <typename EltwiseLayerFunctions> Status validate_eltwise_Layer(EltwiseLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating EltwiseLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating EltwiseLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 2); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -678,23 +706,24 @@ Status validate_eltwise_Layer(EltwiseLayerNode &node) const QuantizationInfo quant_info = node.output_quant_info(); // Validate function - if(eltwise_op == EltwiseOperation::Add) + if (eltwise_op == EltwiseOperation::Add) { return EltwiseLayerFunctions::ArithmeticAddition::validate(input1, input2, output, convert_policy, act_info); } - else if(eltwise_op == EltwiseOperation::Sub) + else if (eltwise_op == EltwiseOperation::Sub) { return EltwiseLayerFunctions::ArithmeticSubtraction::validate(input1, input2, output, convert_policy, act_info); } - else if(eltwise_op == EltwiseOperation::Mul) + else if (eltwise_op == EltwiseOperation::Mul) { - return EltwiseLayerFunctions::PixelWiseMultiplication::validate(input1, input2, output, 1.0f, convert_policy, round_policy, act_info); + return EltwiseLayerFunctions::PixelWiseMultiplication::validate(input1, input2, output, 1.0f, convert_policy, + round_policy, act_info); } - else if(eltwise_op == EltwiseOperation::Max) + else if (eltwise_op == EltwiseOperation::Max) { return EltwiseLayerFunctions::ElementwiseMax::validate(input1, input2, output, act_info); } - else if(eltwise_op == EltwiseOperation::Div) + else if (eltwise_op == EltwiseOperation::Div) { return EltwiseLayerFunctions::ArithmeticDivision::validate(input1, input2, output, act_info); } @@ -713,7 +742,8 @@ Status validate_eltwise_Layer(EltwiseLayerNode &node) template <typename UnaryEltwiseLayerFunctions> Status validate_unary_eltwise_layer(UnaryEltwiseLayerNode &node) { - ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating EltwiseLayer node with ID : " << node.id() << " and Name: " << node.name() << std::endl); + ARM_COMPUTE_LOG_GRAPH_VERBOSE("Validating EltwiseLayer node with ID : " << node.id() << " and Name: " << node.name() + << std::endl); ARM_COMPUTE_RETURN_ERROR_ON(node.num_inputs() != 1); ARM_COMPUTE_RETURN_ERROR_ON(node.num_outputs() != 1); @@ -723,7 +753,7 @@ Status validate_unary_eltwise_layer(UnaryEltwiseLayerNode &node) const UnaryEltwiseOperation eltwise_op = node.eltwise_descriptor().op; // Validate function - if(eltwise_op == UnaryEltwiseOperation::Exp) + if (eltwise_op == UnaryEltwiseOperation::Exp) { return UnaryEltwiseLayerFunctions::ExpLayer::validate(input, output); } diff --git a/arm_compute/graph/frontend/IStream.h b/arm_compute/graph/frontend/IStream.h index f69d5437c1..1831ac0be3 100644 --- a/arm_compute/graph/frontend/IStream.h +++ b/arm_compute/graph/frontend/IStream.h @@ -84,8 +84,8 @@ public: } protected: - StreamHints _hints = {}; /**< Execution and algorithmic hints */ - NodeID _tail_node = { EmptyNodeID }; /**< NodeID pointing to the last(tail) node of the graph */ + StreamHints _hints = {}; /**< Execution and algorithmic hints */ + NodeID _tail_node = {EmptyNodeID}; /**< NodeID pointing to the last(tail) node of the graph */ }; } // namespace frontend } // namespace graph diff --git a/arm_compute/graph/frontend/Layers.h b/arm_compute/graph/frontend/Layers.h index fe0539bac5..bd321e6f1a 100644 --- a/arm_compute/graph/frontend/Layers.h +++ b/arm_compute/graph/frontend/Layers.h @@ -24,13 +24,12 @@ #ifndef ARM_COMPUTE_GRAPH_LAYERS_H #define ARM_COMPUTE_GRAPH_LAYERS_H -#include "arm_compute/graph/GraphBuilder.h" -#include "arm_compute/graph/Types.h" +#include "arm_compute/core/utils/misc/Utility.h" #include "arm_compute/graph/frontend/ILayer.h" #include "arm_compute/graph/frontend/IStream.h" #include "arm_compute/graph/frontend/SubStream.h" - -#include "arm_compute/core/utils/misc/Utility.h" +#include "arm_compute/graph/GraphBuilder.h" +#include "arm_compute/graph/Types.h" #include <memory> #include <string> @@ -50,14 +49,13 @@ public: * @param[in] desc Description of input tensor. * @param[in] accessor Accessor to get input tensor data from. */ - InputLayer(TensorDescriptor desc, ITensorAccessorUPtr accessor) - : _desc(desc), _accessor(std::move(accessor)) + InputLayer(TensorDescriptor desc, ITensorAccessorUPtr accessor) : _desc(desc), _accessor(std::move(accessor)) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; + NodeParams common_params = {name(), s.hints().target_hint}; return GraphBuilder::add_input_node(s.graph(), common_params, _desc, std::move(_accessor)); } @@ -75,14 +73,13 @@ public: * @param[in] desc Description of input tensor. * @param[in] accessor Accessor to get input tensor data from. */ - ConstantLayer(TensorDescriptor desc, ITensorAccessorUPtr accessor) - : _desc(desc), _accessor(std::move(accessor)) + ConstantLayer(TensorDescriptor desc, ITensorAccessorUPtr accessor) : _desc(desc), _accessor(std::move(accessor)) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; + NodeParams common_params = {name(), s.hints().target_hint}; return GraphBuilder::add_const_node(s.graph(), common_params, _desc, std::move(_accessor)); } @@ -107,8 +104,8 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), _connection_idx }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), _connection_idx}; return GraphBuilder::add_output_node(s.graph(), common_params, input, std::move(_accessor)); } @@ -126,18 +123,17 @@ public: * @param[in] act_info Activation information * @param[in] out_quant_info (Optional) Output quantization info */ - ActivationLayer(ActivationLayerInfo act_info, - const QuantizationInfo out_quant_info = QuantizationInfo()) - : _act_info(act_info), - _out_quant_info(std::move(out_quant_info)) + ActivationLayer(ActivationLayerInfo act_info, const QuantizationInfo out_quant_info = QuantizationInfo()) + : _act_info(act_info), _out_quant_info(std::move(out_quant_info)) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; - return GraphBuilder::add_activation_node(s.graph(), common_params, input, _act_info, std::move(_out_quant_info)); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; + return GraphBuilder::add_activation_node(s.graph(), common_params, input, _act_info, + std::move(_out_quant_info)); } private: @@ -160,10 +156,7 @@ public: unsigned int axis, DataType out_data_type = DataType::UNKNOWN, const QuantizationInfo out_quant_info = QuantizationInfo()) - : _op(op), - _axis(axis), - _out_data_type(out_data_type), - _out_quant_info(std::move(out_quant_info)) + : _op(op), _axis(axis), _out_data_type(out_data_type), _out_quant_info(std::move(out_quant_info)) { } @@ -175,9 +168,10 @@ public: */ NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; - return GraphBuilder::add_arg_min_max_node(s.graph(), common_params, input, _op, _axis, _out_data_type, std::move(_out_quant_info)); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; + return GraphBuilder::add_arg_min_max_node(s.graph(), common_params, input, _op, _axis, _out_data_type, + std::move(_out_quant_info)); } private: @@ -204,7 +198,11 @@ public: ITensorAccessorUPtr gamma = nullptr, ITensorAccessorUPtr beta = nullptr, float epsilon = 0.001f) - : _mean(std::move(mean)), _var(std::move(var)), _gamma(std::move(gamma)), _beta(std::move(beta)), _epsilon(epsilon) + : _mean(std::move(mean)), + _var(std::move(var)), + _gamma(std::move(gamma)), + _beta(std::move(beta)), + _epsilon(epsilon) { } @@ -213,10 +211,10 @@ public: ARM_COMPUTE_ERROR_ON(_mean == nullptr); ARM_COMPUTE_ERROR_ON(_var == nullptr); - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; - return GraphBuilder::add_batch_normalization_node(s.graph(), common_params, input, _epsilon, - std::move(_mean), std::move(_var), std::move(_beta), std::move(_gamma)); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; + return GraphBuilder::add_batch_normalization_node(s.graph(), common_params, input, _epsilon, std::move(_mean), + std::move(_var), std::move(_beta), std::move(_gamma)); } private: @@ -237,7 +235,9 @@ public: * @param[in] sub_stream_deltas Graph sub-stream for the deltas * @param[in] info Contains BoundingBox operation information described in @ref BoundingBoxTransformInfo. */ - BoundingBoxTransformLayer(SubStream &&sub_stream_input, SubStream &&sub_stream_deltas, BoundingBoxTransformInfo info) + BoundingBoxTransformLayer(SubStream &&sub_stream_input, + SubStream &&sub_stream_deltas, + BoundingBoxTransformInfo info) : _ss_input(sub_stream_input), _ss_deltas(sub_stream_deltas), _bbox_info(info) { } @@ -250,9 +250,9 @@ public: */ NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { _ss_input.tail_node(), 0 }; - NodeIdxPair deltas = { _ss_deltas.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {_ss_input.tail_node(), 0}; + NodeIdxPair deltas = {_ss_deltas.tail_node(), 0}; return GraphBuilder::add_bounding_box_transform_node(s.graph(), common_params, input, deltas, _bbox_info); } @@ -270,15 +270,14 @@ public: * * @param[in] num_groups Number of groups */ - ChannelShuffleLayer(unsigned int num_groups) - : _num_groups(num_groups) + ChannelShuffleLayer(unsigned int num_groups) : _num_groups(num_groups) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_channel_shuffle_node(s.graph(), common_params, input, _num_groups); } @@ -297,17 +296,15 @@ public: * @param[in] rest_sub_streams Rest sub-graph branches */ template <typename... Ts> - ConcatLayer(SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&... rest_sub_streams) + ConcatLayer(SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&...rest_sub_streams) : _sub_streams(), _concat_descriptor(DataLayoutDimension::CHANNEL) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream1))); _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream2))); - utility::for_each([&](SubStream && sub_stream) - { - _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); - }, - std::move(rest_sub_streams)...); + utility::for_each([&](SubStream &&sub_stream) + { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); }, + std::move(rest_sub_streams)...); } /** Construct a concatenation layer * @@ -317,33 +314,33 @@ public: * @param[in] rest_sub_streams Rest sub-graph branches */ template <typename... Ts> - ConcatLayer(descriptors::ConcatLayerDescriptor concat_descriptor, SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&... rest_sub_streams) + ConcatLayer(descriptors::ConcatLayerDescriptor concat_descriptor, + SubStream &&sub_stream1, + SubStream &&sub_stream2, + Ts &&...rest_sub_streams) : _sub_streams(), _concat_descriptor(concat_descriptor) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream1))); _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream2))); - utility::for_each([&](SubStream && sub_stream) - { - _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); - }, - std::move(rest_sub_streams)...); + utility::for_each([&](SubStream &&sub_stream) + { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); }, + std::move(rest_sub_streams)...); } /** Construct a concat layer * * @param[in] sub_stream Sub-stream */ template <typename... Ts> - ConcatLayer(SubStream &&sub_stream) - : _sub_streams(), _concat_descriptor(DataLayoutDimension::CHANNEL) + ConcatLayer(SubStream &&sub_stream) : _sub_streams(), _concat_descriptor(DataLayoutDimension::CHANNEL) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); } NodeID create_layer(IStream &s) override { NodeID nid = EmptyNodeID; - NodeParams common_params = { name(), s.hints().target_hint }; - if(_sub_streams.size() == 1 && _sub_streams.at(0) != nullptr) + NodeParams common_params = {name(), s.hints().target_hint}; + if (_sub_streams.size() == 1 && _sub_streams.at(0) != nullptr) { nid = _sub_streams[0]->tail_node(); } @@ -351,14 +348,14 @@ public: { // Collect tail nodes and concatenate std::vector<NodeIdxPair> nodes; - for(auto &ss : _sub_streams) + for (auto &ss : _sub_streams) { - if(ss && (ss->tail_node() != EmptyNodeID)) + if (ss && (ss->tail_node() != EmptyNodeID)) { const auto tail_node = s.graph().node(ss->tail_node()); - if(tail_node != nullptr && tail_node->type() != NodeType::Output) + if (tail_node != nullptr && tail_node->type() != NodeType::Output) { - nodes.push_back({ ss->tail_node(), 0 }); + nodes.push_back({ss->tail_node(), 0}); } } } @@ -411,12 +408,12 @@ public: NodeID create_layer(IStream &s) override { - NodeIdxPair input = { s.tail_node(), 0 }; - NodeParams common_params = { name(), s.hints().target_hint }; - return GraphBuilder::add_convolution_node(s.graph(), common_params, input, - Size2D(_conv_width, _conv_height), _ofm, _conv_info, _num_groups, - s.hints().convolution_method_hint, s.hints().fast_math_hint, - std::move(_weights), std::move(_bias), std::move(_weights_quant_info), std::move(_out_quant_info)); + NodeIdxPair input = {s.tail_node(), 0}; + NodeParams common_params = {name(), s.hints().target_hint}; + return GraphBuilder::add_convolution_node(s.graph(), common_params, input, Size2D(_conv_width, _conv_height), + _ofm, _conv_info, _num_groups, s.hints().convolution_method_hint, + s.hints().fast_math_hint, std::move(_weights), std::move(_bias), + std::move(_weights_quant_info), std::move(_out_quant_info)); } private: @@ -461,11 +458,10 @@ public: NodeID create_layer(IStream &s) override { - NodeIdxPair input = { s.tail_node(), 0 }; - NodeParams common_params = { name(), s.hints().target_hint }; - return GraphBuilder::add_deconvolution_node(s.graph(), common_params, input, - Size2D(_conv_width, _conv_height), _ofm, _deconv_info, - std::move(_weights), std::move(_bias)); + NodeIdxPair input = {s.tail_node(), 0}; + NodeParams common_params = {name(), s.hints().target_hint}; + return GraphBuilder::add_deconvolution_node(s.graph(), common_params, input, Size2D(_conv_width, _conv_height), + _ofm, _deconv_info, std::move(_weights), std::move(_bias)); } private: @@ -513,12 +509,12 @@ public: NodeID create_layer(IStream &s) override { - NodeIdxPair input = { s.tail_node(), 0 }; - NodeParams common_params = { name(), s.hints().target_hint }; - return GraphBuilder::add_depthwise_convolution_node(s.graph(), common_params, - input, Size2D(_conv_width, _conv_height), _conv_info, _depth_multiplier, - s.hints().depthwise_convolution_method_hint, - std::move(_weights), std::move(_bias), std::move(_weights_quant_info), std::move(_out_quant_info)); + NodeIdxPair input = {s.tail_node(), 0}; + NodeParams common_params = {name(), s.hints().target_hint}; + return GraphBuilder::add_depthwise_convolution_node( + s.graph(), common_params, input, Size2D(_conv_width, _conv_height), _conv_info, _depth_multiplier, + s.hints().depthwise_convolution_method_hint, std::move(_weights), std::move(_bias), + std::move(_weights_quant_info), std::move(_out_quant_info)); } private: @@ -540,15 +536,14 @@ public: * * @param[in] block_shape Block size to rearranged */ - DepthToSpaceLayer(int32_t block_shape) - : _block_shape(block_shape) + DepthToSpaceLayer(int32_t block_shape) : _block_shape(block_shape) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_depth_to_space_node(s.graph(), common_params, input, _block_shape); } @@ -569,8 +564,8 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_dequantization_node(s.graph(), common_params, input); } }; @@ -585,18 +580,21 @@ public: * @param[in] sub_stream_prior PriorBox graph sub-stream. * @param[in] detect_info DetectionOutput parameters. */ - DetectionOutputLayer(SubStream &&sub_stream_conf, SubStream &&sub_stream_prior, const DetectionOutputLayerInfo &detect_info) + DetectionOutputLayer(SubStream &&sub_stream_conf, + SubStream &&sub_stream_prior, + const DetectionOutputLayerInfo &detect_info) : _ss_conf(std::move(sub_stream_conf)), _ss_prior(std::move(sub_stream_prior)), _detect_info(detect_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input_loc = { s.tail_node(), 0 }; - NodeIdxPair input_conf = { _ss_conf.tail_node(), 0 }; - NodeIdxPair input_priorbox = { _ss_prior.tail_node(), 0 }; - return GraphBuilder::add_detection_output_node(s.graph(), common_params, input_loc, input_conf, input_priorbox, _detect_info); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input_loc = {s.tail_node(), 0}; + NodeIdxPair input_conf = {_ss_conf.tail_node(), 0}; + NodeIdxPair input_priorbox = {_ss_prior.tail_node(), 0}; + return GraphBuilder::add_detection_output_node(s.graph(), common_params, input_loc, input_conf, input_priorbox, + _detect_info); } private: @@ -615,9 +613,14 @@ public: * @param[in] anchors Accessor to get anchors tensor data from. * @param[in] out_quant_info (Optional) Output quantization info */ - DetectionPostProcessLayer(SubStream &&sub_stream_class_prediction, DetectionPostProcessLayerInfo detect_info, ITensorAccessorUPtr anchors, - const QuantizationInfo out_quant_info = QuantizationInfo()) - : _sub_stream_class_prediction(std::move(sub_stream_class_prediction)), _detect_info(detect_info), _anchors(std::move(anchors)), _out_quant_info(std::move(out_quant_info)) + DetectionPostProcessLayer(SubStream &&sub_stream_class_prediction, + DetectionPostProcessLayerInfo detect_info, + ITensorAccessorUPtr anchors, + const QuantizationInfo out_quant_info = QuantizationInfo()) + : _sub_stream_class_prediction(std::move(sub_stream_class_prediction)), + _detect_info(detect_info), + _anchors(std::move(anchors)), + _out_quant_info(std::move(out_quant_info)) { } @@ -625,10 +628,12 @@ public: { ARM_COMPUTE_ERROR_ON(_anchors == nullptr); - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input_box_encoding = { s.tail_node(), 0 }; - NodeIdxPair input_class_prediction = { _sub_stream_class_prediction.tail_node(), 0 }; - return GraphBuilder::add_detection_post_process_node(s.graph(), common_params, input_box_encoding, input_class_prediction, _detect_info, std::move(_anchors), std::move(_out_quant_info)); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input_box_encoding = {s.tail_node(), 0}; + NodeIdxPair input_class_prediction = {_sub_stream_class_prediction.tail_node(), 0}; + return GraphBuilder::add_detection_post_process_node(s.graph(), common_params, input_box_encoding, + input_class_prediction, _detect_info, std::move(_anchors), + std::move(_out_quant_info)); } private: @@ -645,15 +650,14 @@ public: * * @param[in] shape Output shape */ - DummyLayer(TensorShape shape) - : _shape(shape) + DummyLayer(TensorShape shape) : _shape(shape) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_dummy_node(s.graph(), common_params, input, _shape); } @@ -677,9 +681,9 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input0 = { _ss0.tail_node(), 0 }; - NodeIdxPair input1 = { _ss1.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input0 = {_ss0.tail_node(), 0}; + NodeIdxPair input1 = {_ss1.tail_node(), 0}; return GraphBuilder::add_elementwise_node(s.graph(), common_params, input0, input1, _op); } @@ -700,8 +704,8 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_flatten_node(s.graph(), common_params, input); } }; @@ -770,13 +774,13 @@ public: */ NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; - if(_weights != nullptr) + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; + if (_weights != nullptr) { - return GraphBuilder::add_fully_connected_layer(s.graph(), common_params, input, _num_outputs, - std::move(_weights), std::move(_bias), _fc_info, - std::move(_weights_quant_info), std::move(_out_quant_info), s.hints().fast_math_hint); + return GraphBuilder::add_fully_connected_layer( + s.graph(), common_params, input, _num_outputs, std::move(_weights), std::move(_bias), _fc_info, + std::move(_weights_quant_info), std::move(_out_quant_info), s.hints().fast_math_hint); } else { @@ -811,8 +815,14 @@ public: * @param[in] ss_anchors Graph sub-stream for the anchors. * @param[in] info Generate Proposals operation information. */ - GenerateProposalsLayer(SubStream &&ss_scores, SubStream &&ss_deltas, SubStream &&ss_anchors, GenerateProposalsInfo info) - : _ss_scores(std::move(ss_scores)), _ss_deltas(std::move(ss_deltas)), _ss_anchors(std::move(ss_anchors)), _info(info) + GenerateProposalsLayer(SubStream &&ss_scores, + SubStream &&ss_deltas, + SubStream &&ss_anchors, + GenerateProposalsInfo info) + : _ss_scores(std::move(ss_scores)), + _ss_deltas(std::move(ss_deltas)), + _ss_anchors(std::move(ss_anchors)), + _info(info) { } @@ -824,10 +834,10 @@ public: */ NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair scores = { _ss_scores.tail_node(), 0 }; - NodeIdxPair deltas = { _ss_deltas.tail_node(), 0 }; - NodeIdxPair anchors = { _ss_anchors.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair scores = {_ss_scores.tail_node(), 0}; + NodeIdxPair deltas = {_ss_deltas.tail_node(), 0}; + NodeIdxPair anchors = {_ss_anchors.tail_node(), 0}; return GraphBuilder::add_generate_proposals_node(s.graph(), common_params, scores, deltas, anchors, _info); } @@ -847,15 +857,14 @@ public: * @param[in] axis Axis to perform normalization on * @param[in] epsilon Lower bound value for the normalization */ - L2NormalizeLayer(int axis, float epsilon) - : _axis(axis), _epsilon(epsilon) + L2NormalizeLayer(int axis, float epsilon) : _axis(axis), _epsilon(epsilon) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_l2_normalize_node(s.graph(), common_params, input, _axis, _epsilon); } @@ -872,15 +881,14 @@ public: * * @param[in] norm_info Normalization information. */ - NormalizationLayer(NormalizationLayerInfo norm_info) - : _norm_info(norm_info) + NormalizationLayer(NormalizationLayerInfo norm_info) : _norm_info(norm_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_normalization_node(s.graph(), common_params, input, _norm_info); } @@ -897,8 +905,7 @@ public: * @param[in] mean Accessor to get mean tensor data from. * @param[in] std Accessor to get std tensor data from. */ - NormalizePlanarYUVLayer(ITensorAccessorUPtr mean, - ITensorAccessorUPtr std) + NormalizePlanarYUVLayer(ITensorAccessorUPtr mean, ITensorAccessorUPtr std) : _mean(std::move(mean)), _std(std::move(std)) { } @@ -908,10 +915,10 @@ public: ARM_COMPUTE_ERROR_ON(_mean == nullptr); ARM_COMPUTE_ERROR_ON(_std == nullptr); - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; - return GraphBuilder::add_normalize_planar_yuv_node(s.graph(), common_params, input, - std::move(_mean), std::move(_std)); + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; + return GraphBuilder::add_normalize_planar_yuv_node(s.graph(), common_params, input, std::move(_mean), + std::move(_std)); } private: @@ -929,15 +936,14 @@ public: * specifies the front and the end padding in the i-th dimension. * @param[in] pad_value Padding value to use. Defaults to 0. */ - PadLayer(PaddingList padding, PixelValue pad_value = PixelValue()) - : _padding(padding), _pad_value(pad_value) + PadLayer(PaddingList padding, PixelValue pad_value = PixelValue()) : _padding(padding), _pad_value(pad_value) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_pad_node(s.graph(), common_params, input, _padding, _pad_value); } @@ -956,15 +962,14 @@ public: * @param[in] layout (Optional) Data layout to assign to permuted tensor. * If UNKNOWN then the input's layout will be used. */ - PermuteLayer(PermutationVector perm, DataLayout layout = DataLayout::UNKNOWN) - : _perm(perm), _layout(layout) + PermuteLayer(PermutationVector perm, DataLayout layout = DataLayout::UNKNOWN) : _perm(perm), _layout(layout) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_permute_node(s.graph(), common_params, input, _perm, _layout); } @@ -981,15 +986,14 @@ public: * * @param[in] pool_info Pooling information. */ - PoolingLayer(PoolingLayerInfo pool_info) - : _pool_info(pool_info) + PoolingLayer(PoolingLayerInfo pool_info) : _pool_info(pool_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_pooling_node(s.graph(), common_params, input, _pool_info); } @@ -1013,9 +1017,9 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { _ss0.tail_node(), 0 }; - NodeIdxPair alpha = { _ss1.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {_ss0.tail_node(), 0}; + NodeIdxPair alpha = {_ss1.tail_node(), 0}; return GraphBuilder::add_prelu_node(s.graph(), common_params, input, alpha); } @@ -1064,15 +1068,17 @@ public: * @param[in] format_info (Optional) Format info. * @param[in] transform (Optional) Input transform function. */ - PrintLayer(std::ostream &stream, const IOFormatInfo &format_info = IOFormatInfo(), const std::function<ITensor *(ITensor *)> transform = nullptr) + PrintLayer(std::ostream &stream, + const IOFormatInfo &format_info = IOFormatInfo(), + const std::function<ITensor *(ITensor *)> transform = nullptr) : _stream(stream), _format_info(format_info), _transform(transform) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_print_node(s.graph(), common_params, input, _stream, _format_info, _transform); } @@ -1098,9 +1104,9 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input0 = { s.tail_node(), 0 }; - NodeIdxPair input1 = { _ss.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input0 = {s.tail_node(), 0}; + NodeIdxPair input1 = {_ss.tail_node(), 0}; return GraphBuilder::add_priorbox_node(s.graph(), common_params, input0, input1, _prior_info); } @@ -1117,15 +1123,14 @@ public: * * @param[in] out_quant_info Output tensor quantization info */ - QuantizationLayer(QuantizationInfo out_quant_info) - : _out_quant_info(out_quant_info) + QuantizationLayer(QuantizationInfo out_quant_info) : _out_quant_info(out_quant_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_quantization_node(s.graph(), common_params, input, _out_quant_info); } @@ -1150,8 +1155,8 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_reduction_operation_node(s.graph(), common_params, input, _op, _axis, _keep_dims); } @@ -1170,15 +1175,14 @@ public: * @param[in] stride Stride value to use for reorganizing the values in the output tensor. * It defines the spatial distance between 2 consecutive pixels in the x and y direction */ - ReorgLayer(int stride) - : _stride(stride) + ReorgLayer(int stride) : _stride(stride) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_reorg_node(s.graph(), common_params, input, _stride); } @@ -1194,15 +1198,14 @@ public: * * @param[in] shape Target shape. */ - ReshapeLayer(TensorShape shape) - : _shape(shape) + ReshapeLayer(TensorShape shape) : _shape(shape) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_reshape_node(s.graph(), common_params, input, _shape); } @@ -1221,8 +1224,8 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_resize_node(s.graph(), common_params, input, _policy, _width_scale, _height_scale); } @@ -1254,9 +1257,9 @@ public: NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { _ss_input.tail_node(), 0 }; - NodeIdxPair rois = { _ss_rois.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {_ss_input.tail_node(), 0}; + NodeIdxPair rois = {_ss_rois.tail_node(), 0}; return GraphBuilder::add_roi_align_node(s.graph(), common_params, input, rois, _pool_info); } @@ -1275,16 +1278,15 @@ public: * @param[in] mul_w Accessor to get mul weight from. * @param[in] add_w Accessor to get add weight from. */ - ScaleLayer(ITensorAccessorUPtr mul_w, - ITensorAccessorUPtr add_w) + ScaleLayer(ITensorAccessorUPtr mul_w, ITensorAccessorUPtr add_w) : _mul_w(std::move(mul_w)), _add_w(std::move(add_w)) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_scale_layer(s.graph(), common_params, input, std::move(_mul_w), std::move(_add_w)); } @@ -1302,15 +1304,14 @@ public: * @param[in] starts The starts of the dimensions of the input tensor to be sliced. The length must be of rank(input). * @param[in] ends The ends of the dimensions of the input tensor to be sliced. The length must be of rank(input). */ - SliceLayer(Coordinates &starts, Coordinates &ends) - : _starts(starts), _ends(ends) + SliceLayer(Coordinates &starts, Coordinates &ends) : _starts(starts), _ends(ends) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_slice_node(s.graph(), common_params, input, _starts, _ends); } @@ -1327,15 +1328,14 @@ public: * * @param[in] beta (Optional) Beta value. Default 1.0. */ - SoftmaxLayer(float beta = 1.0f) - : _beta(beta) + SoftmaxLayer(float beta = 1.0f) : _beta(beta) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_softmax_node(s.graph(), common_params, input, _beta); } @@ -1354,17 +1354,14 @@ public: * @param[in] rest_sub_streams Rest sub-graph branches */ template <typename... Ts> - StackLayer(SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&... rest_sub_streams) - : _sub_streams(), _axis(0) + StackLayer(SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&...rest_sub_streams) : _sub_streams(), _axis(0) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream1))); _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream2))); - utility::for_each([&](SubStream && sub_stream) - { - _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); - }, - std::move(rest_sub_streams)...); + utility::for_each([&](SubStream &&sub_stream) + { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); }, + std::move(rest_sub_streams)...); } /** Construct a concatenation layer * @@ -1374,33 +1371,30 @@ public: * @param[in] rest_sub_streams Rest sub-graph branches */ template <typename... Ts> - StackLayer(int axis, SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&... rest_sub_streams) + StackLayer(int axis, SubStream &&sub_stream1, SubStream &&sub_stream2, Ts &&...rest_sub_streams) : _sub_streams(), _axis(axis) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream1))); _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream2))); - utility::for_each([&](SubStream && sub_stream) - { - _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); - }, - std::move(rest_sub_streams)...); + utility::for_each([&](SubStream &&sub_stream) + { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); }, + std::move(rest_sub_streams)...); } /** Construct a concat layer * * @param[in] sub_stream Sub-stream */ template <typename... Ts> - StackLayer(SubStream &&sub_stream) - : _sub_streams(), _axis(0) + StackLayer(SubStream &&sub_stream) : _sub_streams(), _axis(0) { _sub_streams.push_back(std::make_unique<SubStream>(std::move(sub_stream))); } NodeID create_layer(IStream &s) override { NodeID nid = EmptyNodeID; - NodeParams common_params = { name(), s.hints().target_hint }; - if(_sub_streams.size() == 1 && _sub_streams.at(0) != nullptr) + NodeParams common_params = {name(), s.hints().target_hint}; + if (_sub_streams.size() == 1 && _sub_streams.at(0) != nullptr) { nid = _sub_streams[0]->tail_node(); } @@ -1408,14 +1402,14 @@ public: { // Collect tail nodes and stack std::vector<NodeIdxPair> nodes; - for(auto &ss : _sub_streams) + for (auto &ss : _sub_streams) { - if(ss && (ss->tail_node() != EmptyNodeID)) + if (ss && (ss->tail_node() != EmptyNodeID)) { const auto tail_node = s.graph().node(ss->tail_node()); - if(tail_node != nullptr && tail_node->type() != NodeType::Output) + if (tail_node != nullptr && tail_node->type() != NodeType::Output) { - nodes.push_back({ ss->tail_node(), 0 }); + nodes.push_back({ss->tail_node(), 0}); } } } @@ -1440,15 +1434,18 @@ public: * @param[in] strides The strides of the dimensions of the input tensor to be sliced. The length must be of rank(input). * @param[in] strided_slice_info Contains masks for the starts, ends and strides */ - StridedSliceLayer(Coordinates &starts, Coordinates &ends, BiStrides &strides, StridedSliceLayerInfo strided_slice_info) + StridedSliceLayer(Coordinates &starts, + Coordinates &ends, + BiStrides &strides, + StridedSliceLayerInfo strided_slice_info) : _starts(starts), _ends(ends), _strides(strides), _info(strided_slice_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_strided_slice_node(s.graph(), common_params, input, _starts, _ends, _strides, _info); } @@ -1467,15 +1464,14 @@ public: * * @param[in] act_info Activation info */ - YOLOLayer(ActivationLayerInfo act_info) - : _act_info(act_info) + YOLOLayer(ActivationLayerInfo act_info) : _act_info(act_info) { } NodeID create_layer(IStream &s) override { - NodeParams common_params = { name(), s.hints().target_hint }; - NodeIdxPair input = { s.tail_node(), 0 }; + NodeParams common_params = {name(), s.hints().target_hint}; + NodeIdxPair input = {s.tail_node(), 0}; return GraphBuilder::add_yolo_node(s.graph(), common_params, input, _act_info); } diff --git a/arm_compute/graph/frontend/Stream.h b/arm_compute/graph/frontend/Stream.h index db22f6d91b..7e760b6373 100644 --- a/arm_compute/graph/frontend/Stream.h +++ b/arm_compute/graph/frontend/Stream.h @@ -27,7 +27,6 @@ #include "arm_compute/graph/frontend/IStream.h" #include "arm_compute/graph/frontend/IStreamOperators.h" #include "arm_compute/graph/frontend/Types.h" - #include "arm_compute/graph/Graph.h" #include "arm_compute/graph/GraphContext.h" #include "arm_compute/graph/GraphManager.h" @@ -65,7 +64,7 @@ public: void run(); // Inherited overridden methods - void add_layer(ILayer &layer) override; + void add_layer(ILayer &layer) override; Graph &graph() override; const Graph &graph() const override; diff --git a/arm_compute/graph/frontend/SubStream.h b/arm_compute/graph/frontend/SubStream.h index 2283cfeebe..c54317c52b 100644 --- a/arm_compute/graph/frontend/SubStream.h +++ b/arm_compute/graph/frontend/SubStream.h @@ -54,7 +54,7 @@ public: SubStream(IStream &s); // Inherited overridden methods - void add_layer(ILayer &layer) override; + void add_layer(ILayer &layer) override; Graph &graph() override; const Graph &graph() const override; diff --git a/arm_compute/graph/frontend/Types.h b/arm_compute/graph/frontend/Types.h index bc4fe7ae0d..42b28b3cd2 100644 --- a/arm_compute/graph/frontend/Types.h +++ b/arm_compute/graph/frontend/Types.h @@ -33,39 +33,40 @@ namespace graph namespace frontend { // Import types for graph -using graph::DataType; using graph::DataLayout; using graph::DataLayoutDimension; -using graph::TensorShape; +using graph::DataType; using graph::PermutationVector; +using graph::TensorShape; using graph::ActivationLayerInfo; +using graph::ConvolutionMethod; +using graph::DepthwiseConvolutionMethod; +using graph::DimensionRoundingType; using graph::EltwiseOperation; +using graph::FastMathHint; using graph::FullyConnectedLayerInfo; +using graph::GraphConfig; +using graph::InterpolationPolicy; using graph::NormalizationLayerInfo; using graph::NormType; using graph::PadStrideInfo; using graph::PoolingLayerInfo; using graph::PoolingType; +using graph::Size2D; using graph::Target; -using graph::ConvolutionMethod; -using graph::FastMathHint; -using graph::DepthwiseConvolutionMethod; using graph::TensorDescriptor; -using graph::DimensionRoundingType; -using graph::GraphConfig; -using graph::InterpolationPolicy; -using graph::Size2D; /** Hints that can be passed to the stream to expose parameterization */ struct StreamHints { - Target target_hint = { Target::UNSPECIFIED }; /**< Target execution hint */ - ConvolutionMethod convolution_method_hint = { ConvolutionMethod::Default }; /**< Convolution method hint */ - DepthwiseConvolutionMethod depthwise_convolution_method_hint = { DepthwiseConvolutionMethod::Default }; /**< Depthwise Convolution method hint */ - FastMathHint fast_math_hint = { FastMathHint::Disabled }; /**< Fast math hint */ + Target target_hint = {Target::UNSPECIFIED}; /**< Target execution hint */ + ConvolutionMethod convolution_method_hint = {ConvolutionMethod::Default}; /**< Convolution method hint */ + DepthwiseConvolutionMethod depthwise_convolution_method_hint = { + DepthwiseConvolutionMethod::Default}; /**< Depthwise Convolution method hint */ + FastMathHint fast_math_hint = {FastMathHint::Disabled}; /**< Fast math hint */ }; } // namespace frontend } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_GRAPH_STREAM_TYPES_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_GRAPH_STREAM_TYPES_H */ diff --git a/arm_compute/graph/mutators/DepthConcatSubTensorMutator.h b/arm_compute/graph/mutators/DepthConcatSubTensorMutator.h index cb1f079a2e..61d8854a61 100644 --- a/arm_compute/graph/mutators/DepthConcatSubTensorMutator.h +++ b/arm_compute/graph/mutators/DepthConcatSubTensorMutator.h @@ -40,7 +40,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/GroupedConvolutionMutator.h b/arm_compute/graph/mutators/GroupedConvolutionMutator.h index e68c7030d0..3ed8d786fc 100644 --- a/arm_compute/graph/mutators/GroupedConvolutionMutator.h +++ b/arm_compute/graph/mutators/GroupedConvolutionMutator.h @@ -40,7 +40,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/InPlaceOperationMutator.h b/arm_compute/graph/mutators/InPlaceOperationMutator.h index 6248d86a0a..86f62f1994 100644 --- a/arm_compute/graph/mutators/InPlaceOperationMutator.h +++ b/arm_compute/graph/mutators/InPlaceOperationMutator.h @@ -37,7 +37,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/NodeExecutionMethodMutator.h b/arm_compute/graph/mutators/NodeExecutionMethodMutator.h index 07c8ffad97..505d4ab300 100644 --- a/arm_compute/graph/mutators/NodeExecutionMethodMutator.h +++ b/arm_compute/graph/mutators/NodeExecutionMethodMutator.h @@ -42,7 +42,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/NodeFusionMutator.h b/arm_compute/graph/mutators/NodeFusionMutator.h index f3e3eaa190..9d2d44f436 100644 --- a/arm_compute/graph/mutators/NodeFusionMutator.h +++ b/arm_compute/graph/mutators/NodeFusionMutator.h @@ -38,7 +38,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/SplitLayerSubTensorMutator.h b/arm_compute/graph/mutators/SplitLayerSubTensorMutator.h index b14ef59532..ab9746a29b 100644 --- a/arm_compute/graph/mutators/SplitLayerSubTensorMutator.h +++ b/arm_compute/graph/mutators/SplitLayerSubTensorMutator.h @@ -40,7 +40,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/mutators/SyntheticDataTypeMutator.h b/arm_compute/graph/mutators/SyntheticDataTypeMutator.h index 2292e52086..ce8af0a1d7 100644 --- a/arm_compute/graph/mutators/SyntheticDataTypeMutator.h +++ b/arm_compute/graph/mutators/SyntheticDataTypeMutator.h @@ -40,7 +40,7 @@ public: // Inherited methods overridden virtual void mutate(Graph &g) override; MutationType type() const override; - const char *name() override; + const char *name() override; private: DataType _mutate_type; diff --git a/arm_compute/graph/nodes/ActivationLayerNode.h b/arm_compute/graph/nodes/ActivationLayerNode.h index 4a98ee248f..fe5f273db5 100644 --- a/arm_compute/graph/nodes/ActivationLayerNode.h +++ b/arm_compute/graph/nodes/ActivationLayerNode.h @@ -39,8 +39,7 @@ public: * @param[in] info Activation Layer information * @param[in] out_quant_info (Optional) Output quantization info */ - ActivationLayerNode(ActivationLayerInfo info, - QuantizationInfo out_quant_info = QuantizationInfo()); + ActivationLayerNode(ActivationLayerInfo info, QuantizationInfo out_quant_info = QuantizationInfo()); /** Activation metadata accessor * * @return The activation info of the layer @@ -51,7 +50,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::ActivationLayer; diff --git a/arm_compute/graph/nodes/ArgMinMaxLayerNode.h b/arm_compute/graph/nodes/ArgMinMaxLayerNode.h index 69191add99..65fbc36db6 100644 --- a/arm_compute/graph/nodes/ArgMinMaxLayerNode.h +++ b/arm_compute/graph/nodes/ArgMinMaxLayerNode.h @@ -65,7 +65,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::ArgMinMaxLayer; diff --git a/arm_compute/graph/nodes/BatchNormalizationLayerNode.h b/arm_compute/graph/nodes/BatchNormalizationLayerNode.h index e7f4049df8..8583ed87eb 100644 --- a/arm_compute/graph/nodes/BatchNormalizationLayerNode.h +++ b/arm_compute/graph/nodes/BatchNormalizationLayerNode.h @@ -60,7 +60,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::BatchNormalizationLayer; diff --git a/arm_compute/graph/nodes/BoundingBoxTransformLayerNode.h b/arm_compute/graph/nodes/BoundingBoxTransformLayerNode.h index 57175eba2e..96c2544065 100644 --- a/arm_compute/graph/nodes/BoundingBoxTransformLayerNode.h +++ b/arm_compute/graph/nodes/BoundingBoxTransformLayerNode.h @@ -50,7 +50,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: BoundingBoxTransformInfo _bbox_info; diff --git a/arm_compute/graph/nodes/ChannelShuffleLayerNode.h b/arm_compute/graph/nodes/ChannelShuffleLayerNode.h index 0696fe56fc..d296a2dcc3 100644 --- a/arm_compute/graph/nodes/ChannelShuffleLayerNode.h +++ b/arm_compute/graph/nodes/ChannelShuffleLayerNode.h @@ -49,7 +49,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: unsigned int _num_groups; diff --git a/arm_compute/graph/nodes/ConcatenateLayerNode.h b/arm_compute/graph/nodes/ConcatenateLayerNode.h index 8582403355..13398b1a61 100644 --- a/arm_compute/graph/nodes/ConcatenateLayerNode.h +++ b/arm_compute/graph/nodes/ConcatenateLayerNode.h @@ -47,7 +47,8 @@ public: * * @return Expected output descriptor */ - static TensorDescriptor compute_output_descriptor(const std::vector<TensorDescriptor> &input_descriptors, DataLayoutDimension axis); + static TensorDescriptor compute_output_descriptor(const std::vector<TensorDescriptor> &input_descriptors, + DataLayoutDimension axis); /** Disables or not the depth concatenate node * * @warning This is used when concatenate is performed using sub-tensors, where this node is used as a placeholder. @@ -78,7 +79,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: unsigned int _total_nodes; diff --git a/arm_compute/graph/nodes/ConstNode.h b/arm_compute/graph/nodes/ConstNode.h index b377c60208..400b9b4d9f 100644 --- a/arm_compute/graph/nodes/ConstNode.h +++ b/arm_compute/graph/nodes/ConstNode.h @@ -44,7 +44,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: TensorDescriptor _desc; diff --git a/arm_compute/graph/nodes/ConvolutionLayerNode.h b/arm_compute/graph/nodes/ConvolutionLayerNode.h index 99effa07dc..8a77b89f27 100644 --- a/arm_compute/graph/nodes/ConvolutionLayerNode.h +++ b/arm_compute/graph/nodes/ConvolutionLayerNode.h @@ -111,7 +111,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::ConvolutionLayer; diff --git a/arm_compute/graph/nodes/DeconvolutionLayerNode.h b/arm_compute/graph/nodes/DeconvolutionLayerNode.h index e74adb17aa..553d05985c 100644 --- a/arm_compute/graph/nodes/DeconvolutionLayerNode.h +++ b/arm_compute/graph/nodes/DeconvolutionLayerNode.h @@ -61,7 +61,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: descriptors::DeconvolutionLayerDescriptor descriptor; diff --git a/arm_compute/graph/nodes/DepthToSpaceLayerNode.h b/arm_compute/graph/nodes/DepthToSpaceLayerNode.h index 25e30e2c67..5fbcc670ff 100644 --- a/arm_compute/graph/nodes/DepthToSpaceLayerNode.h +++ b/arm_compute/graph/nodes/DepthToSpaceLayerNode.h @@ -56,7 +56,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: int _block_shape; diff --git a/arm_compute/graph/nodes/DepthwiseConvolutionLayerNode.h b/arm_compute/graph/nodes/DepthwiseConvolutionLayerNode.h index 5df86983f0..441d68d2b8 100644 --- a/arm_compute/graph/nodes/DepthwiseConvolutionLayerNode.h +++ b/arm_compute/graph/nodes/DepthwiseConvolutionLayerNode.h @@ -101,7 +101,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::DepthwiseConvolutionLayer; diff --git a/arm_compute/graph/nodes/DequantizationLayerNode.h b/arm_compute/graph/nodes/DequantizationLayerNode.h index 4910938d47..1cce71373f 100644 --- a/arm_compute/graph/nodes/DequantizationLayerNode.h +++ b/arm_compute/graph/nodes/DequantizationLayerNode.h @@ -46,8 +46,8 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; }; } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_GRAPH_DEQUANTIZATION_NODE_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_GRAPH_DEQUANTIZATION_NODE_H */ diff --git a/arm_compute/graph/nodes/DetectionOutputLayerNode.h b/arm_compute/graph/nodes/DetectionOutputLayerNode.h index b4b910c40e..c3e067e430 100644 --- a/arm_compute/graph/nodes/DetectionOutputLayerNode.h +++ b/arm_compute/graph/nodes/DetectionOutputLayerNode.h @@ -51,13 +51,14 @@ public: * * @return Output descriptor */ - static TensorDescriptor compute_output_descriptor(const TensorDescriptor &input_descriptor, const DetectionOutputLayerInfo &info); + static TensorDescriptor compute_output_descriptor(const TensorDescriptor &input_descriptor, + const DetectionOutputLayerInfo &info); // Inherited overridden methods: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: DetectionOutputLayerInfo _info; diff --git a/arm_compute/graph/nodes/DetectionPostProcessLayerNode.h b/arm_compute/graph/nodes/DetectionPostProcessLayerNode.h index 6ff78aee07..a53aaf2b9c 100644 --- a/arm_compute/graph/nodes/DetectionPostProcessLayerNode.h +++ b/arm_compute/graph/nodes/DetectionPostProcessLayerNode.h @@ -49,7 +49,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: DetectionPostProcessLayerInfo _info; @@ -59,4 +59,4 @@ private: }; } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_GRAPH_DETECTION_POST_PROCESS_LAYER_NODE_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_GRAPH_DETECTION_POST_PROCESS_LAYER_NODE_H */ diff --git a/arm_compute/graph/nodes/DummyNode.h b/arm_compute/graph/nodes/DummyNode.h index 645f1b325d..2263525a72 100644 --- a/arm_compute/graph/nodes/DummyNode.h +++ b/arm_compute/graph/nodes/DummyNode.h @@ -51,11 +51,11 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: TensorShape _shape; }; } // namespace graph } // namespace arm_compute -#endif /* ARM_COMPUTE_GRAPH_DUMMY_NODE_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_GRAPH_DUMMY_NODE_H */ diff --git a/arm_compute/graph/nodes/EltwiseLayerNode.h b/arm_compute/graph/nodes/EltwiseLayerNode.h index 7a6d8e8303..258298259f 100644 --- a/arm_compute/graph/nodes/EltwiseLayerNode.h +++ b/arm_compute/graph/nodes/EltwiseLayerNode.h @@ -79,7 +79,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; static constexpr NodeType node_type = NodeType::EltwiseLayer; @@ -112,7 +112,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; static constexpr NodeType node_type = NodeType::UnaryEltwiseLayer; diff --git a/arm_compute/graph/nodes/FlattenLayerNode.h b/arm_compute/graph/nodes/FlattenLayerNode.h index 046114c291..af104707a1 100644 --- a/arm_compute/graph/nodes/FlattenLayerNode.h +++ b/arm_compute/graph/nodes/FlattenLayerNode.h @@ -41,7 +41,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/nodes/FullyConnectedLayerNode.h b/arm_compute/graph/nodes/FullyConnectedLayerNode.h index 9ade62bf4a..3bcf386d64 100644 --- a/arm_compute/graph/nodes/FullyConnectedLayerNode.h +++ b/arm_compute/graph/nodes/FullyConnectedLayerNode.h @@ -73,7 +73,7 @@ public: */ static TensorDescriptor compute_weights_descriptor(const TensorDescriptor &input_descriptor, unsigned int num_outputs, - FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), + FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), const QuantizationInfo &weights_quant_info = QuantizationInfo()); /** Computes fully connected layer output descriptor * @@ -98,7 +98,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; static constexpr NodeType node_type = NodeType::FullyConnectedLayer; diff --git a/arm_compute/graph/nodes/FusedConvolutionBatchNormalizationNode.h b/arm_compute/graph/nodes/FusedConvolutionBatchNormalizationNode.h index b0051b1385..d891ea49eb 100644 --- a/arm_compute/graph/nodes/FusedConvolutionBatchNormalizationNode.h +++ b/arm_compute/graph/nodes/FusedConvolutionBatchNormalizationNode.h @@ -43,7 +43,8 @@ public: * @param[in] fast_math_hint (Optional) Fast math hint * @param[in] fused_activation (Optional) Fused activation layer. Disabled if not specified */ - FusedConvolutionBatchNormalizationNode(float epsilon, PadStrideInfo info, + FusedConvolutionBatchNormalizationNode(float epsilon, + PadStrideInfo info, unsigned int num_groups = 1, ConvolutionMethod method = ConvolutionMethod::Default, FastMathHint fast_math_hint = FastMathHint::Disabled, @@ -122,7 +123,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::FusedConvolutionBatchNormalizationLayer; diff --git a/arm_compute/graph/nodes/FusedDepthwiseConvolutionBatchNormalizationNode.h b/arm_compute/graph/nodes/FusedDepthwiseConvolutionBatchNormalizationNode.h index a01cb9dc42..a61b155151 100644 --- a/arm_compute/graph/nodes/FusedDepthwiseConvolutionBatchNormalizationNode.h +++ b/arm_compute/graph/nodes/FusedDepthwiseConvolutionBatchNormalizationNode.h @@ -46,7 +46,7 @@ public: PadStrideInfo info, unsigned int depth_multiplier, DepthwiseConvolutionMethod method, - ActivationLayerInfo fused_activation = ActivationLayerInfo()); + ActivationLayerInfo fused_activation = ActivationLayerInfo()); /** Sets the depthwise convolution layer method to use * @@ -117,7 +117,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::FusedDepthwiseConvolutionBatchNormalizationLayer; diff --git a/arm_compute/graph/nodes/GenerateProposalsLayerNode.h b/arm_compute/graph/nodes/GenerateProposalsLayerNode.h index 6f8edc8758..b5e4b9781c 100644 --- a/arm_compute/graph/nodes/GenerateProposalsLayerNode.h +++ b/arm_compute/graph/nodes/GenerateProposalsLayerNode.h @@ -50,7 +50,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: GenerateProposalsInfo _info; diff --git a/arm_compute/graph/nodes/InputNode.h b/arm_compute/graph/nodes/InputNode.h index 07091af64f..0983d25a59 100644 --- a/arm_compute/graph/nodes/InputNode.h +++ b/arm_compute/graph/nodes/InputNode.h @@ -44,7 +44,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: TensorDescriptor _desc; diff --git a/arm_compute/graph/nodes/L2NormalizeLayerNode.h b/arm_compute/graph/nodes/L2NormalizeLayerNode.h index 8edc5b0bf3..ed11412b70 100644 --- a/arm_compute/graph/nodes/L2NormalizeLayerNode.h +++ b/arm_compute/graph/nodes/L2NormalizeLayerNode.h @@ -68,7 +68,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: int _axis; diff --git a/arm_compute/graph/nodes/Nodes.h b/arm_compute/graph/nodes/Nodes.h index ae9f177ec4..d4ad32b6f0 100644 --- a/arm_compute/graph/nodes/Nodes.h +++ b/arm_compute/graph/nodes/Nodes.h @@ -50,18 +50,18 @@ #include "arm_compute/graph/nodes/NormalizationLayerNode.h" #include "arm_compute/graph/nodes/NormalizePlanarYUVLayerNode.h" #include "arm_compute/graph/nodes/OutputNode.h" -#include "arm_compute/graph/nodes/PReluLayerNode.h" #include "arm_compute/graph/nodes/PadLayerNode.h" #include "arm_compute/graph/nodes/PermuteLayerNode.h" #include "arm_compute/graph/nodes/PoolingLayerNode.h" +#include "arm_compute/graph/nodes/PReluLayerNode.h" #include "arm_compute/graph/nodes/PrintLayerNode.h" #include "arm_compute/graph/nodes/PriorBoxLayerNode.h" #include "arm_compute/graph/nodes/QuantizationLayerNode.h" -#include "arm_compute/graph/nodes/ROIAlignLayerNode.h" #include "arm_compute/graph/nodes/ReductionLayerNode.h" #include "arm_compute/graph/nodes/ReorgLayerNode.h" #include "arm_compute/graph/nodes/ReshapeLayerNode.h" #include "arm_compute/graph/nodes/ResizeLayerNode.h" +#include "arm_compute/graph/nodes/ROIAlignLayerNode.h" #include "arm_compute/graph/nodes/SliceLayerNode.h" #include "arm_compute/graph/nodes/SoftmaxLayerNode.h" #include "arm_compute/graph/nodes/SplitLayerNode.h" diff --git a/arm_compute/graph/nodes/NormalizationLayerNode.h b/arm_compute/graph/nodes/NormalizationLayerNode.h index 503b859e53..86f2fb9dba 100644 --- a/arm_compute/graph/nodes/NormalizationLayerNode.h +++ b/arm_compute/graph/nodes/NormalizationLayerNode.h @@ -49,7 +49,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: NormalizationLayerInfo _info; diff --git a/arm_compute/graph/nodes/NormalizePlanarYUVLayerNode.h b/arm_compute/graph/nodes/NormalizePlanarYUVLayerNode.h index 4d84c20de0..158acc4c23 100644 --- a/arm_compute/graph/nodes/NormalizePlanarYUVLayerNode.h +++ b/arm_compute/graph/nodes/NormalizePlanarYUVLayerNode.h @@ -41,7 +41,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/nodes/OutputNode.h b/arm_compute/graph/nodes/OutputNode.h index c91bc6b699..75484ab328 100644 --- a/arm_compute/graph/nodes/OutputNode.h +++ b/arm_compute/graph/nodes/OutputNode.h @@ -41,7 +41,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/nodes/PReluLayerNode.h b/arm_compute/graph/nodes/PReluLayerNode.h index b8e6c1ae7f..532fdccb3a 100644 --- a/arm_compute/graph/nodes/PReluLayerNode.h +++ b/arm_compute/graph/nodes/PReluLayerNode.h @@ -41,7 +41,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; }; } // namespace graph } // namespace arm_compute diff --git a/arm_compute/graph/nodes/PadLayerNode.h b/arm_compute/graph/nodes/PadLayerNode.h index d6ff3553da..dcb5ea595b 100644 --- a/arm_compute/graph/nodes/PadLayerNode.h +++ b/arm_compute/graph/nodes/PadLayerNode.h @@ -56,7 +56,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::PadLayer; diff --git a/arm_compute/graph/nodes/PermuteLayerNode.h b/arm_compute/graph/nodes/PermuteLayerNode.h index 0b2380b51c..62654e777c 100644 --- a/arm_compute/graph/nodes/PermuteLayerNode.h +++ b/arm_compute/graph/nodes/PermuteLayerNode.h @@ -51,7 +51,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: PermutationVector _perm; diff --git a/arm_compute/graph/nodes/PoolingLayerNode.h b/arm_compute/graph/nodes/PoolingLayerNode.h index b336bb906f..c81f3f98dc 100644 --- a/arm_compute/graph/nodes/PoolingLayerNode.h +++ b/arm_compute/graph/nodes/PoolingLayerNode.h @@ -57,7 +57,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: PoolingLayerInfo _info; diff --git a/arm_compute/graph/nodes/PrintLayerNode.h b/arm_compute/graph/nodes/PrintLayerNode.h index b57ac1f6d4..e7accc8015 100644 --- a/arm_compute/graph/nodes/PrintLayerNode.h +++ b/arm_compute/graph/nodes/PrintLayerNode.h @@ -43,7 +43,9 @@ public: * @param[in] format_info (Optional) Format info. * @param[in] transform (Optional) Input transform function. */ - PrintLayerNode(std::ostream &stream, const IOFormatInfo &format_info = IOFormatInfo(), const std::function<ITensor *(ITensor *)> transform = nullptr); + PrintLayerNode(std::ostream &stream, + const IOFormatInfo &format_info = IOFormatInfo(), + const std::function<ITensor *(ITensor *)> transform = nullptr); /** Stream metadata accessor * @@ -67,7 +69,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: std::ostream &_stream; diff --git a/arm_compute/graph/nodes/PriorBoxLayerNode.h b/arm_compute/graph/nodes/PriorBoxLayerNode.h index c7eadd1fe5..db36bfb1e0 100644 --- a/arm_compute/graph/nodes/PriorBoxLayerNode.h +++ b/arm_compute/graph/nodes/PriorBoxLayerNode.h @@ -51,13 +51,14 @@ public: * * @return Output descriptor */ - static TensorDescriptor compute_output_descriptor(const TensorDescriptor &input_descriptor, const PriorBoxLayerInfo &info); + static TensorDescriptor compute_output_descriptor(const TensorDescriptor &input_descriptor, + const PriorBoxLayerInfo &info); // Inherited overridden methods: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: PriorBoxLayerInfo _info; diff --git a/arm_compute/graph/nodes/QuantizationLayerNode.h b/arm_compute/graph/nodes/QuantizationLayerNode.h index e5d81afa0e..b8e4c7d27b 100644 --- a/arm_compute/graph/nodes/QuantizationLayerNode.h +++ b/arm_compute/graph/nodes/QuantizationLayerNode.h @@ -51,7 +51,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; static constexpr NodeType node_type = NodeType::QuantizationLayer; diff --git a/arm_compute/graph/nodes/ROIAlignLayerNode.h b/arm_compute/graph/nodes/ROIAlignLayerNode.h index 5abd0659b5..70309a551c 100644 --- a/arm_compute/graph/nodes/ROIAlignLayerNode.h +++ b/arm_compute/graph/nodes/ROIAlignLayerNode.h @@ -56,7 +56,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: ROIPoolingLayerInfo _pool_info; diff --git a/arm_compute/graph/nodes/ReductionLayerNode.h b/arm_compute/graph/nodes/ReductionLayerNode.h index b8d295945c..ff99466c8f 100644 --- a/arm_compute/graph/nodes/ReductionLayerNode.h +++ b/arm_compute/graph/nodes/ReductionLayerNode.h @@ -56,7 +56,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: ReductionOperation _op; diff --git a/arm_compute/graph/nodes/ReorgLayerNode.h b/arm_compute/graph/nodes/ReorgLayerNode.h index 986692ed28..a3bbcdb00f 100644 --- a/arm_compute/graph/nodes/ReorgLayerNode.h +++ b/arm_compute/graph/nodes/ReorgLayerNode.h @@ -57,7 +57,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: int _stride; diff --git a/arm_compute/graph/nodes/ReshapeLayerNode.h b/arm_compute/graph/nodes/ReshapeLayerNode.h index 727d253ce5..992275c2b1 100644 --- a/arm_compute/graph/nodes/ReshapeLayerNode.h +++ b/arm_compute/graph/nodes/ReshapeLayerNode.h @@ -44,7 +44,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: TensorShape _shape; diff --git a/arm_compute/graph/nodes/ResizeLayerNode.h b/arm_compute/graph/nodes/ResizeLayerNode.h index 79f8889f9c..480d6e517f 100644 --- a/arm_compute/graph/nodes/ResizeLayerNode.h +++ b/arm_compute/graph/nodes/ResizeLayerNode.h @@ -51,7 +51,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: InterpolationPolicy _policy; diff --git a/arm_compute/graph/nodes/SliceLayerNode.h b/arm_compute/graph/nodes/SliceLayerNode.h index 08d3794e26..63f266b217 100644 --- a/arm_compute/graph/nodes/SliceLayerNode.h +++ b/arm_compute/graph/nodes/SliceLayerNode.h @@ -51,7 +51,8 @@ public: * @return Output descriptor */ static TensorDescriptor compute_output_descriptor(const TensorDescriptor &input_descriptor, - const Coordinates &starts, const Coordinates &ends); + const Coordinates &starts, + const Coordinates &ends); /** Start coordinates accessor * * @return Start coordinates of the dimensions @@ -67,7 +68,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: Coordinates _starts; diff --git a/arm_compute/graph/nodes/SoftmaxLayerNode.h b/arm_compute/graph/nodes/SoftmaxLayerNode.h index 0868c6ff16..2cb1ac2cf4 100644 --- a/arm_compute/graph/nodes/SoftmaxLayerNode.h +++ b/arm_compute/graph/nodes/SoftmaxLayerNode.h @@ -49,7 +49,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; public: static constexpr NodeType node_type = NodeType::SoftmaxLayer; diff --git a/arm_compute/graph/nodes/SplitLayerNode.h b/arm_compute/graph/nodes/SplitLayerNode.h index 13cccdd447..5e6df53c0f 100644 --- a/arm_compute/graph/nodes/SplitLayerNode.h +++ b/arm_compute/graph/nodes/SplitLayerNode.h @@ -55,7 +55,9 @@ public: * @return A pair with the descriptor of the split and the starting coordinates */ std::pair<TensorDescriptor, Coordinates> compute_output_descriptor(const TensorDescriptor &input_descriptor, - unsigned int num_splits, int axis, unsigned int idx); + unsigned int num_splits, + int axis, + unsigned int idx); /** Number of splits accessor * * @return Number of splits @@ -72,7 +74,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: unsigned int _num_splits; diff --git a/arm_compute/graph/nodes/StackLayerNode.h b/arm_compute/graph/nodes/StackLayerNode.h index 2990895c2b..9f0767c9f2 100644 --- a/arm_compute/graph/nodes/StackLayerNode.h +++ b/arm_compute/graph/nodes/StackLayerNode.h @@ -58,7 +58,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: unsigned int _total_nodes; diff --git a/arm_compute/graph/nodes/StridedSliceLayerNode.h b/arm_compute/graph/nodes/StridedSliceLayerNode.h index 6039f312b3..f521feb780 100644 --- a/arm_compute/graph/nodes/StridedSliceLayerNode.h +++ b/arm_compute/graph/nodes/StridedSliceLayerNode.h @@ -84,7 +84,7 @@ public: NodeType type() const override; bool forward_descriptors() override; TensorDescriptor configure_output(size_t idx) const override; - void accept(INodeVisitor &v) override; + void accept(INodeVisitor &v) override; private: Coordinates _starts; diff --git a/arm_compute/graph/printers/DotGraphPrinter.h b/arm_compute/graph/printers/DotGraphPrinter.h index 564aecfb1e..6638033044 100644 --- a/arm_compute/graph/printers/DotGraphPrinter.h +++ b/arm_compute/graph/printers/DotGraphPrinter.h @@ -25,7 +25,6 @@ #define ACL_ARM_COMPUTE_GRAPH_PRINTERS_DOTGRAPHPRINTER_H #include "arm_compute/graph/IGraphPrinter.h" - #include "arm_compute/graph/INodeVisitor.h" #include <string> diff --git a/arm_compute/runtime/Allocator.h b/arm_compute/runtime/Allocator.h index 83f072ab6b..e99ddb3dac 100644 --- a/arm_compute/runtime/Allocator.h +++ b/arm_compute/runtime/Allocator.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_ALLOCATOR_H #include "arm_compute/runtime/IAllocator.h" - #include "arm_compute/runtime/IMemoryRegion.h" #include <cstddef> @@ -40,9 +39,9 @@ public: Allocator() = default; // Inherited methods overridden: - void *allocate(size_t size, size_t alignment) override; - void free(void *ptr) override; + void *allocate(size_t size, size_t alignment) override; + void free(void *ptr) override; std::unique_ptr<IMemoryRegion> make_region(size_t size, size_t alignment) override; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_ALLOCATOR_H */ diff --git a/arm_compute/runtime/Array.h b/arm_compute/runtime/Array.h index 21d9c25c87..9283273317 100644 --- a/arm_compute/runtime/Array.h +++ b/arm_compute/runtime/Array.h @@ -37,16 +37,14 @@ class Array : public IArray<T> { public: /** Default constructor: empty array */ - Array() - : IArray<T>(0), _values(nullptr) + Array() : IArray<T>(0), _values(nullptr) { } /** Constructor: initializes an array which can contain up to max_num_points values * * @param[in] max_num_values Maximum number of values the array will be able to stored */ - Array(size_t max_num_values) - : IArray<T>(max_num_values), _values(std::make_unique<T[]>(max_num_values)) + Array(size_t max_num_values) : IArray<T>(max_num_values), _values(std::make_unique<T[]>(max_num_values)) { } @@ -72,5 +70,5 @@ using Int16Array = Array<int16_t>; using Int32Array = Array<int32_t>; /** Array of floats. */ using FloatArray = Array<float>; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_ARRAY_H */ diff --git a/arm_compute/runtime/BlobLifetimeManager.h b/arm_compute/runtime/BlobLifetimeManager.h index 0d69f2e7c5..18ffe96ee5 100644 --- a/arm_compute/runtime/BlobLifetimeManager.h +++ b/arm_compute/runtime/BlobLifetimeManager.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_BLOBLIFETIMEMANAGER_H #include "arm_compute/runtime/ISimpleLifetimeManager.h" - #include "arm_compute/runtime/Types.h" #include <memory> @@ -62,7 +61,7 @@ public: // Inherited methods overridden: std::unique_ptr<IMemoryPool> create_pool(IAllocator *allocator) override; - MappingType mapping_type() const override; + MappingType mapping_type() const override; private: // Inherited methods overridden: diff --git a/arm_compute/runtime/BlobMemoryPool.h b/arm_compute/runtime/BlobMemoryPool.h index 8481fa20f9..b25efc3821 100644 --- a/arm_compute/runtime/BlobMemoryPool.h +++ b/arm_compute/runtime/BlobMemoryPool.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_BLOBMEMORYPOOL_H #include "arm_compute/runtime/IMemoryPool.h" - #include "arm_compute/runtime/IMemoryRegion.h" #include "arm_compute/runtime/Types.h" @@ -62,8 +61,8 @@ public: BlobMemoryPool &operator=(BlobMemoryPool &&) = default; // Inherited methods overridden: - void acquire(MemoryMappings &handles) override; - void release(MemoryMappings &handles) override; + void acquire(MemoryMappings &handles) override; + void release(MemoryMappings &handles) override; MappingType mapping_type() const override; std::unique_ptr<IMemoryPool> duplicate() override; diff --git a/arm_compute/runtime/CL/CLArray.h b/arm_compute/runtime/CL/CLArray.h index 7efe208b9f..6e81a46a29 100644 --- a/arm_compute/runtime/CL/CLArray.h +++ b/arm_compute/runtime/CL/CLArray.h @@ -38,8 +38,7 @@ class CLArray : public ICLArray<T> { public: /** Default constructor: empty array */ - CLArray() - : ICLArray<T>(0), _buffer() + CLArray() : ICLArray<T>(0), _buffer() { } /** Prevent instances of this class from being copied (As this class contains pointers) */ @@ -55,7 +54,8 @@ public: * @param[in] max_num_values Maximum number of values the array will be able to stored */ CLArray(size_t max_num_values) - : ICLArray<T>(max_num_values), _buffer(CLScheduler::get().context(), CL_MEM_ALLOC_HOST_PTR | CL_MEM_READ_WRITE, max_num_values * sizeof(T)) + : ICLArray<T>(max_num_values), + _buffer(CLScheduler::get().context(), CL_MEM_ALLOC_HOST_PTR | CL_MEM_READ_WRITE, max_num_values * sizeof(T)) { } /** Enqueue a map operation of the allocated buffer. @@ -91,7 +91,8 @@ protected: uint8_t *do_map(cl::CommandQueue &q, bool blocking) override { ARM_COMPUTE_ERROR_ON(nullptr == _buffer.get()); - return static_cast<uint8_t *>(q.enqueueMapBuffer(_buffer, blocking ? CL_TRUE : CL_FALSE, CL_MAP_READ | CL_MAP_WRITE, 0, this->max_num_values() * sizeof(T))); + return static_cast<uint8_t *>(q.enqueueMapBuffer( + _buffer, blocking ? CL_TRUE : CL_FALSE, CL_MAP_READ | CL_MAP_WRITE, 0, this->max_num_values() * sizeof(T))); } void do_unmap(cl::CommandQueue &q, uint8_t *mapping) override { @@ -114,5 +115,5 @@ using CLInt16Array = CLArray<cl_short>; using CLInt32Array = CLArray<cl_int>; /** OpenCL Array of floats. */ using CLFloatArray = CLArray<cl_float>; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLARRAY_H */ diff --git a/arm_compute/runtime/CL/CLBufferAllocator.h b/arm_compute/runtime/CL/CLBufferAllocator.h index 7467e9d1c6..00ff017012 100644 --- a/arm_compute/runtime/CL/CLBufferAllocator.h +++ b/arm_compute/runtime/CL/CLBufferAllocator.h @@ -35,9 +35,9 @@ class CLBufferAllocator final : public IAllocator { public: // Inherited methods overridden: - void *allocate(size_t size, size_t alignment) override; - void free(void *ptr) override; + void *allocate(size_t size, size_t alignment) override; + void free(void *ptr) override; std::unique_ptr<IMemoryRegion> make_region(size_t size, size_t alignment) override; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_CLBUFFERALLOCATOR_H */ diff --git a/arm_compute/runtime/CL/CLMemory.h b/arm_compute/runtime/CL/CLMemory.h index 7adee66c73..5abe86bd53 100644 --- a/arm_compute/runtime/CL/CLMemory.h +++ b/arm_compute/runtime/CL/CLMemory.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_RUNTIME_CL_CLMEMORY_H #define ARM_COMPUTE_RUNTIME_CL_CLMEMORY_H -#include "arm_compute/runtime/IMemory.h" - #include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/runtime/CL/CLMemoryRegion.h" +#include "arm_compute/runtime/IMemory.h" #include <cstddef> #include <memory> @@ -75,8 +74,8 @@ public: // Inherited methods overridden: IMemoryRegion *region() final; IMemoryRegion *region() const final; - void set_region(IMemoryRegion *region) final; - void set_owned_region(std::unique_ptr<IMemoryRegion> region) final; + void set_region(IMemoryRegion *region) final; + void set_owned_region(std::unique_ptr<IMemoryRegion> region) final; private: ICLMemoryRegion *_region; diff --git a/arm_compute/runtime/CL/CLMemoryRegion.h b/arm_compute/runtime/CL/CLMemoryRegion.h index 66a30fa56b..365973a9e6 100644 --- a/arm_compute/runtime/CL/CLMemoryRegion.h +++ b/arm_compute/runtime/CL/CLMemoryRegion.h @@ -110,7 +110,7 @@ public: // Inherited methods overridden : void *ptr() final; void *map(cl::CommandQueue &q, bool blocking) final; - void unmap(cl::CommandQueue &q) final; + void unmap(cl::CommandQueue &q) final; }; /** OpenCL SVM memory region interface */ @@ -156,7 +156,7 @@ public: // Inherited methods overridden : void *map(cl::CommandQueue &q, bool blocking) final; - void unmap(cl::CommandQueue &q) final; + void unmap(cl::CommandQueue &q) final; }; /** OpenCL fine-grain SVM memory region implementation */ @@ -173,7 +173,7 @@ public: // Inherited methods overridden : void *map(cl::CommandQueue &q, bool blocking) final; - void unmap(cl::CommandQueue &q) final; + void unmap(cl::CommandQueue &q) final; }; } // namespace arm_compute #endif /* ARM_COMPUTE_RUNTIME_CL_CL_MEMORY_REGION_H */ diff --git a/arm_compute/runtime/CL/CLRuntimeContext.h b/arm_compute/runtime/CL/CLRuntimeContext.h index dd17645fa7..2ed4b74796 100644 --- a/arm_compute/runtime/CL/CLRuntimeContext.h +++ b/arm_compute/runtime/CL/CLRuntimeContext.h @@ -54,11 +54,11 @@ public: CLKernelLibrary &kernel_library(); private: - std::unique_ptr<CLScheduler> _gpu_owned_scheduler{ nullptr }; - CLScheduler *_gpu_scheduler{ nullptr }; - CLTuner _tuner{ false }; + std::unique_ptr<CLScheduler> _gpu_owned_scheduler{nullptr}; + CLScheduler *_gpu_scheduler{nullptr}; + CLTuner _tuner{false}; CLSymbols _symbols{}; - CLBackendType _backend_type{ CLBackendType::Native }; + CLBackendType _backend_type{CLBackendType::Native}; }; } // namespace arm_compute #endif /*ARM_COMPUTE_CLRUNTIME_CONTEXT_H */ diff --git a/arm_compute/runtime/CL/CLScheduler.h b/arm_compute/runtime/CL/CLScheduler.h index 3030239270..b74fcb74ef 100644 --- a/arm_compute/runtime/CL/CLScheduler.h +++ b/arm_compute/runtime/CL/CLScheduler.h @@ -28,8 +28,8 @@ #include "arm_compute/core/CL/CLTypes.h" #include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/core/Error.h" -#include "arm_compute/core/Types.h" #include "arm_compute/core/experimental/Types.h" +#include "arm_compute/core/Types.h" #include "arm_compute/runtime/CL/CLGEMMHeuristicsHandle.h" #include "arm_compute/runtime/CL/CLHelpers.h" #include "arm_compute/runtime/CL/CLTypes.h" @@ -63,7 +63,9 @@ public: * @param[in] gemm_h (Optional) Pointer to CLGEMMHeuristicsHandle (default = nullptr) * @param[in] cl_backend_type (Optional) Type of backend to use (default = CLBackendType::Native) */ - void default_init(ICLTuner *cl_tuner = nullptr, CLGEMMHeuristicsHandle *gemm_h = nullptr, CLBackendType cl_backend_type = CLBackendType::Native); + void default_init(ICLTuner *cl_tuner = nullptr, + CLGEMMHeuristicsHandle *gemm_h = nullptr, + CLBackendType cl_backend_type = CLBackendType::Native); /** Initialises the scheduler with context and device provided by the user * * @param[in] device OpenCL device to be used @@ -71,7 +73,10 @@ public: * @param[in] cl_tuner (Optional) Pointer to ICLTuner (default=nullptr) * @param[in] gemm_h (Optional) Pointer to CLGEMMHeuristicsHandle (default = nullptr) */ - void default_init_with_context(cl::Device &device, cl::Context &ctx, ICLTuner *cl_tuner = nullptr, CLGEMMHeuristicsHandle *gemm_h = nullptr); + void default_init_with_context(cl::Device &device, + cl::Context &ctx, + ICLTuner *cl_tuner = nullptr, + CLGEMMHeuristicsHandle *gemm_h = nullptr); /** Re-initializes the context and command queue used by the scheduler to default values * and sets a default device and kernel path for the @ref CLKernelLibrary. @@ -80,7 +85,9 @@ public: * @param[in] gemm_h (Optional) Pointer to CLGEMMHeuristicsHandle (default = nullptr) * @param[in] cl_backend_type (Optional) Type of backend to use (default = CLBackendType::Native) */ - void default_reinit(ICLTuner *cl_tuner = nullptr, CLGEMMHeuristicsHandle *gemm_h = nullptr, CLBackendType cl_backend_type = CLBackendType::Native); + void default_reinit(ICLTuner *cl_tuner = nullptr, + CLGEMMHeuristicsHandle *gemm_h = nullptr, + CLBackendType cl_backend_type = CLBackendType::Native); /** Schedule the execution of the passed kernel if possible. * @@ -105,8 +112,12 @@ public: * @param[in] gemm_h (Optional) Pointer to CLGEMMHeuristicsHandle (default = nullptr) * @param[in] cl_backend_type (Optional) Type of backend to use (default = CLBackendType::Native) */ - void init(cl::Context context, cl::CommandQueue queue, const cl::Device &device, ICLTuner *cl_tuner = nullptr, CLGEMMHeuristicsHandle *gemm_h = nullptr, - CLBackendType cl_backend_type = CLBackendType::Native); + void init(cl::Context context, + cl::CommandQueue queue, + const cl::Device &device, + ICLTuner *cl_tuner = nullptr, + CLGEMMHeuristicsHandle *gemm_h = nullptr, + CLBackendType cl_backend_type = CLBackendType::Native); /** Accessor for the associated CL context. * diff --git a/arm_compute/runtime/CL/CLSubTensor.h b/arm_compute/runtime/CL/CLSubTensor.h index 0a7f5f89b2..c18df8086a 100644 --- a/arm_compute/runtime/CL/CLSubTensor.h +++ b/arm_compute/runtime/CL/CLSubTensor.h @@ -46,7 +46,10 @@ public: * @param[in] coords Coordinates of the first subtensor element inside the parent tensor. * @param[in] extend_parent (Optional) Extend parent with subtensor shape if subtensor indexes out of bounds */ - CLSubTensor(ICLTensor *parent, const TensorShape &tensor_shape, const Coordinates &coords, bool extend_parent = false); + CLSubTensor(ICLTensor *parent, + const TensorShape &tensor_shape, + const Coordinates &coords, + bool extend_parent = false); /** Destructor: free the tensor's memory */ ~CLSubTensor() = default; /** Restrict instances of this class to be copy constructed */ @@ -93,11 +96,11 @@ public: protected: // Inherited methods overridden: uint8_t *do_map(cl::CommandQueue &q, bool blocking) override; - void do_unmap(cl::CommandQueue &q) override; + void do_unmap(cl::CommandQueue &q) override; private: ICLTensor *_parent; mutable SubTensorInfo _info; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CLSUBTENSOR_H */ diff --git a/arm_compute/runtime/CL/CLTensor.h b/arm_compute/runtime/CL/CLTensor.h index ae73351f27..0729935e9e 100644 --- a/arm_compute/runtime/CL/CLTensor.h +++ b/arm_compute/runtime/CL/CLTensor.h @@ -87,17 +87,17 @@ public: TensorInfo *info() override; const cl::Buffer &cl_buffer() const override; CLQuantization quantization() const override; - void associate_memory_group(IMemoryGroup *memory_group) override; + void associate_memory_group(IMemoryGroup *memory_group) override; CLRuntimeContext *context(); protected: // Inherited methods overridden: uint8_t *do_map(cl::CommandQueue &q, bool blocking) override; - void do_unmap(cl::CommandQueue &q) override; + void do_unmap(cl::CommandQueue &q) override; private: mutable CLTensorAllocator _allocator; /**< Instance of the OpenCL tensor allocator */ - CLRuntimeContext *_ctx{ nullptr }; + CLRuntimeContext *_ctx{nullptr}; }; /** OpenCL Image */ diff --git a/arm_compute/runtime/CL/CLTensorAllocator.h b/arm_compute/runtime/CL/CLTensorAllocator.h index 1b061ee1d6..fde8e9c43a 100644 --- a/arm_compute/runtime/CL/CLTensorAllocator.h +++ b/arm_compute/runtime/CL/CLTensorAllocator.h @@ -24,15 +24,14 @@ #ifndef ARM_COMPUTE_CLTENSORALLOCATOR_H #define ARM_COMPUTE_CLTENSORALLOCATOR_H +#include "arm_compute/core/CL/CLTypes.h" +#include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/runtime/CL/CLArray.h" #include "arm_compute/runtime/CL/CLMemory.h" #include "arm_compute/runtime/IAllocator.h" #include "arm_compute/runtime/ITensorAllocator.h" #include "arm_compute/runtime/MemoryGroup.h" -#include "arm_compute/core/CL/CLTypes.h" -#include "arm_compute/core/CL/OpenCL.h" - #include <cstdint> namespace arm_compute @@ -148,7 +147,7 @@ private: static const cl::Buffer _empty_buffer; private: - CLRuntimeContext *_ctx; + CLRuntimeContext *_ctx; IMemoryManageable *_owner; /**< Memory manageable object that owns the allocator */ IMemoryGroup *_associated_memory_group; /**< Registered memory manager */ CLMemory _memory; /**< OpenCL memory */ diff --git a/arm_compute/runtime/CL/CLTuner.h b/arm_compute/runtime/CL/CLTuner.h index 93aa45adc1..cf293d3d27 100644 --- a/arm_compute/runtime/CL/CLTuner.h +++ b/arm_compute/runtime/CL/CLTuner.h @@ -153,9 +153,9 @@ private: std::unordered_map<std::string, CLTuningParams> _tuning_params_table; std::unordered_map<std::string, cl::NDRange> _lws_table; - cl::Event _kernel_event; - bool _tune_new_kernels; - CLTuningInfo _tuning_info; + cl::Event _kernel_event; + bool _tune_new_kernels; + CLTuningInfo _tuning_info; }; } // namespace arm_compute #endif /*ARM_COMPUTE_CLTUNER_H */ diff --git a/arm_compute/runtime/CL/CLTunerTypes.h b/arm_compute/runtime/CL/CLTunerTypes.h index 508cafac95..d9b914676a 100644 --- a/arm_compute/runtime/CL/CLTunerTypes.h +++ b/arm_compute/runtime/CL/CLTunerTypes.h @@ -43,7 +43,7 @@ enum class CLTunerMode struct CLTuningInfo { CLTunerMode tuner_mode = CLTunerMode::NORMAL; /**< Parameter to select the level (granularity) of the tuning */ - bool tune_wbsm = false; /**< Flag to tune the batches of work groups distributed to compute units. + bool tune_wbsm = false; /**< Flag to tune the batches of work groups distributed to compute units. Internally, the library will check if this feature is available on the target platform. This OpenCL tuner extension is still in experimental phase */ }; @@ -56,11 +56,10 @@ struct CLTuningInfo */ inline CLTunerMode tuner_mode_from_name(const std::string &name) { - static const std::map<std::string, CLTunerMode> tuner_modes = - { - { "exhaustive", CLTunerMode::EXHAUSTIVE }, - { "normal", CLTunerMode::NORMAL }, - { "rapid", CLTunerMode::RAPID }, + static const std::map<std::string, CLTunerMode> tuner_modes = { + {"exhaustive", CLTunerMode::EXHAUSTIVE}, + {"normal", CLTunerMode::NORMAL}, + {"rapid", CLTunerMode::RAPID}, }; #ifndef ARM_COMPUTE_EXCEPTIONS_DISABLED @@ -71,7 +70,7 @@ inline CLTunerMode tuner_mode_from_name(const std::string &name) #ifndef ARM_COMPUTE_EXCEPTIONS_DISABLED } - catch(const std::out_of_range &) + catch (const std::out_of_range &) { throw std::invalid_argument(name); } diff --git a/arm_compute/runtime/CL/CLTuningParams.h b/arm_compute/runtime/CL/CLTuningParams.h index 1e5ab25c26..a876fad112 100644 --- a/arm_compute/runtime/CL/CLTuningParams.h +++ b/arm_compute/runtime/CL/CLTuningParams.h @@ -26,6 +26,7 @@ #include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/runtime/CL/CLTunerTypes.h" + #include "support/StringSupport.h" #include <ostream> @@ -36,8 +37,7 @@ namespace arm_compute class CLTuningParams { public: - CLTuningParams(const CLTuningParams &tuning_params) - : _lws(tuning_params._lws), _wbsm(tuning_params._wbsm) + CLTuningParams(const CLTuningParams &tuning_params) : _lws(tuning_params._lws), _wbsm(tuning_params._wbsm) { } @@ -45,18 +45,16 @@ public: : _lws(lws_x, lws_y, lws_z), _wbsm(wbsm) { } - CLTuningParams(cl::NDRange lws, cl_int wbsm = 0) - : _lws(lws), _wbsm(wbsm) + CLTuningParams(cl::NDRange lws, cl_int wbsm = 0) : _lws(lws), _wbsm(wbsm) { } - CLTuningParams(cl_int wbsm) - : CLTuningParams(cl::NullRange, wbsm) + CLTuningParams(cl_int wbsm) : CLTuningParams(cl::NullRange, wbsm) { } - CLTuningParams& operator=(const CLTuningParams &other) + CLTuningParams &operator=(const CLTuningParams &other) { - _lws = other._lws; + _lws = other._lws; _wbsm = other._wbsm; return *this; } @@ -84,8 +82,9 @@ public: std::string to_string(CLTuningInfo tuning_info) { std::string tuning_params_string = ""; - tuning_params_string += ";" + support::cpp11::to_string(_lws[0]) + ";" + support::cpp11::to_string(_lws[1]) + ";" + support::cpp11::to_string(_lws[2]); - if(tuning_info.tune_wbsm) + tuning_params_string += ";" + support::cpp11::to_string(_lws[0]) + ";" + support::cpp11::to_string(_lws[1]) + + ";" + support::cpp11::to_string(_lws[2]); + if (tuning_info.tune_wbsm) { tuning_params_string += ";" + support::cpp11::to_string(_wbsm); } @@ -98,19 +97,19 @@ public: std::vector<std::string> array; std::stringstream ss(tuning_params_string); std::string temp; - while(ss >> temp) + while (ss >> temp) { array.push_back(temp); } // Read 3 values for lws - if(array.size() < 3) + if (array.size() < 3) { return false; } const unsigned int lws_0 = support::cpp11::stoi(array[0]); const unsigned int lws_1 = support::cpp11::stoi(array[1]); const unsigned int lws_2 = support::cpp11::stoi(array[2]); - if(lws_0 == 0 && lws_1 == 0 && lws_2 == 0) + if (lws_0 == 0 && lws_1 == 0 && lws_2 == 0) { // If lws values are 0, cl::NullRange has to be used // otherwise the lws object will be badly created @@ -121,9 +120,9 @@ public: _lws = cl::NDRange(lws_0, lws_1, lws_2); } array.erase(array.begin(), array.begin() + 3); - if(tuning_info.tune_wbsm) + if (tuning_info.tune_wbsm) { - if(array.size() < 1) + if (array.size() < 1) { return false; } diff --git a/arm_compute/runtime/CL/CLTypes.h b/arm_compute/runtime/CL/CLTypes.h index d298ecd614..931740c47f 100644 --- a/arm_compute/runtime/CL/CLTypes.h +++ b/arm_compute/runtime/CL/CLTypes.h @@ -43,12 +43,12 @@ enum class CLGEMMKernelType /** OpenCL GEMM kernel selection parameters. These information are retrieved to select the GEMM kernel on OpenCL */ struct CLGEMMKernelSelectionParams { - unsigned int m{ 0 }; /**< Number of rows for the lhs matrix. Lhs matrix NOT transposed */ - unsigned int n{ 0 }; /**< Number of columns for the rhs matrix. Rhs matrix NOT transposed */ - unsigned int k{ 0 }; /**< Number of rows for the rhs matrix. Rhs matrix NOT transposed */ - unsigned int b{ 0 }; /**< Batch size */ - bool is_rhs_constant{ false }; /**< True if the content of the rhs matrix is constant */ - DataType data_type{ DataType::UNKNOWN }; /**< Data type */ + unsigned int m{0}; /**< Number of rows for the lhs matrix. Lhs matrix NOT transposed */ + unsigned int n{0}; /**< Number of columns for the rhs matrix. Rhs matrix NOT transposed */ + unsigned int k{0}; /**< Number of rows for the rhs matrix. Rhs matrix NOT transposed */ + unsigned int b{0}; /**< Batch size */ + bool is_rhs_constant{false}; /**< True if the content of the rhs matrix is constant */ + DataType data_type{DataType::UNKNOWN}; /**< Data type */ }; /** List the possible OpenCL backends */ diff --git a/arm_compute/runtime/CL/ICLGEMMKernelSelection.h b/arm_compute/runtime/CL/ICLGEMMKernelSelection.h index 7be9393388..5a71a61203 100644 --- a/arm_compute/runtime/CL/ICLGEMMKernelSelection.h +++ b/arm_compute/runtime/CL/ICLGEMMKernelSelection.h @@ -40,8 +40,7 @@ public: * * @param[in] arch GPU target */ - ICLGEMMKernelSelection(GPUTarget arch) - : _target(arch) + ICLGEMMKernelSelection(GPUTarget arch) : _target(arch) { } /** Default Move Constructor. */ @@ -59,7 +58,8 @@ public: virtual CLGEMMKernelType select_kernel(const CLGEMMKernelSelectionParams ¶ms) = 0; protected: - GPUTarget _target; /**< GPU target could be used to call a dedicated heuristic for each GPU IP for a given GPU architecture */ + GPUTarget + _target; /**< GPU target could be used to call a dedicated heuristic for each GPU IP for a given GPU architecture */ }; } // namespace cl_gemm } // namespace arm_compute diff --git a/arm_compute/runtime/CL/ICLOperator.h b/arm_compute/runtime/CL/ICLOperator.h index 38bcaf32f2..c0826e7733 100644 --- a/arm_compute/runtime/CL/ICLOperator.h +++ b/arm_compute/runtime/CL/ICLOperator.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_ICLOPERATOR_H #include "arm_compute/core/Types.h" - #include "arm_compute/runtime/IOperator.h" #include "arm_compute/runtime/IRuntimeContext.h" #include "arm_compute/runtime/Types.h" @@ -56,8 +55,8 @@ public: ICLOperator &operator=(ICLOperator &&) = default; // Inherited methods overridden: - void run(ITensorPack &tensors) override; - void prepare(ITensorPack &constants) override; + void run(ITensorPack &tensors) override; + void prepare(ITensorPack &constants) override; MemoryRequirements workspace() const override; protected: diff --git a/arm_compute/runtime/CL/functions/CLActivationLayer.h b/arm_compute/runtime/CL/functions/CLActivationLayer.h index 4a718ab4b6..e158efa093 100644 --- a/arm_compute/runtime/CL/functions/CLActivationLayer.h +++ b/arm_compute/runtime/CL/functions/CLActivationLayer.h @@ -24,11 +24,10 @@ #ifndef ARM_COMPUTE_CLACTIVATIONLAYER_H #define ARM_COMPUTE_CLACTIVATIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" #include "arm_compute/runtime/CL/CLRuntimeContext.h" +#include "arm_compute/runtime/IFunction.h" namespace arm_compute { @@ -91,7 +90,10 @@ public: * @param[out] output Destination tensor. Data type supported: same as @p input * @param[in] act_info Activation layer parameters. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, ActivationLayerInfo act_info); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + ActivationLayerInfo act_info); /** Static function to check if given info will lead to a valid configuration of @ref CLActivationLayer * * @param[in] input Source tensor info. In case of @p output tensor info = nullptr, this tensor will store the result diff --git a/arm_compute/runtime/CL/functions/CLArgMinMaxLayer.h b/arm_compute/runtime/CL/functions/CLArgMinMaxLayer.h index ce5bee8d95..d340d20a1f 100644 --- a/arm_compute/runtime/CL/functions/CLArgMinMaxLayer.h +++ b/arm_compute/runtime/CL/functions/CLArgMinMaxLayer.h @@ -91,7 +91,11 @@ public: * @param[out] output Output source tensor. Data types supported: U32/S32. * @param[in] op Reduction operation to perform. Operations supported: ARG_IDX_MAX, ARG_IDX_MIN */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, int axis, ICLTensor *output, const ReductionOperation &op); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + int axis, + ICLTensor *output, + const ReductionOperation &op); /** Static function to check if given info will lead to a valid configuration of @ref CLArgMinMaxLayer * * @param[in] input Input source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/S32/F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLBatchNormalizationLayer.h b/arm_compute/runtime/CL/functions/CLBatchNormalizationLayer.h index 37a0680709..f57bc8fe8b 100644 --- a/arm_compute/runtime/CL/functions/CLBatchNormalizationLayer.h +++ b/arm_compute/runtime/CL/functions/CLBatchNormalizationLayer.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_CLBATCHNORMALIZATIONLAYER_H #define ARM_COMPUTE_CLBATCHNORMALIZATIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -84,7 +83,13 @@ public: * @param[in] epsilon (Optional) Small value to avoid division with zero. Default value is 0.001f. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported. */ - void configure(ICLTensor *input, ICLTensor *output, const ICLTensor *mean, const ICLTensor *var, const ICLTensor *beta = nullptr, const ICLTensor *gamma = nullptr, float epsilon = 0.001f, + void configure(ICLTensor *input, + ICLTensor *output, + const ICLTensor *mean, + const ICLTensor *var, + const ICLTensor *beta = nullptr, + const ICLTensor *gamma = nullptr, + float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo()); /** Set the input and output tensors. * @@ -102,9 +107,15 @@ public: * @param[in] epsilon (Optional) Small value to avoid division with zero. Default value is 0.001f. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const ICLTensor *mean, const ICLTensor *var, const ICLTensor *beta = nullptr, - const ICLTensor *gamma = nullptr, - float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + const ICLTensor *mean, + const ICLTensor *var, + const ICLTensor *beta = nullptr, + const ICLTensor *gamma = nullptr, + float epsilon = 0.001f, + ActivationLayerInfo act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLBatchNormalizationLayer * * @param[in] input Source tensor info. In case of @p output tensor info = nullptr, this tensor will store the result. @@ -120,10 +131,14 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const ITensorInfo *mean, const ITensorInfo *var, - const ITensorInfo *beta = nullptr, const ITensorInfo *gamma = nullptr, - float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const ITensorInfo *mean, + const ITensorInfo *var, + const ITensorInfo *beta = nullptr, + const ITensorInfo *gamma = nullptr, + float epsilon = 0.001f, + ActivationLayerInfo act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLBatchToSpaceLayer.h b/arm_compute/runtime/CL/functions/CLBatchToSpaceLayer.h index 861330b9d4..20b9fdafed 100644 --- a/arm_compute/runtime/CL/functions/CLBatchToSpaceLayer.h +++ b/arm_compute/runtime/CL/functions/CLBatchToSpaceLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLBATCHTOSPACELAYER_H #define ARM_COMPUTE_CLBATCHTOSPACELAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -82,7 +81,10 @@ public: * @deprecated This method for dynamic block shape is not fully mature and will be removed in 23.08 release */ ARM_COMPUTE_DEPRECATED_REL(23.05) - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *block_shape, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *block_shape, + ICLTensor *output); /** Set the input and output tensors. (Static block shape). * * @param[in] input Tensor input. Supported tensor rank: 4. Data types supported: All. @@ -91,7 +93,11 @@ public: * @param[out] output Tensor output. Data types supported: same as @p input * @param[in] crop_info Information about how the output shape is cropped after batch to space is performed */ - void configure(const ICLTensor *input, int32_t block_shape_x, int32_t block_shape_y, ICLTensor *output, const CropInfo &crop_info = CropInfo{}); + void configure(const ICLTensor *input, + int32_t block_shape_x, + int32_t block_shape_y, + ICLTensor *output, + const CropInfo &crop_info = CropInfo{}); /** Set the input and output tensors. (Static block shape). * * @param[in] compile_context The compile context to be used. @@ -101,7 +107,12 @@ public: * @param[out] output Tensor output. Data types supported: same as @p input * @param[in] crop_info Information about how the output shape is cropped after batch to space is performed */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, int32_t block_shape_x, int32_t block_shape_y, ICLTensor *output, const CropInfo &crop_info = CropInfo{}); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + int32_t block_shape_x, + int32_t block_shape_y, + ICLTensor *output, + const CropInfo &crop_info = CropInfo{}); /** Static function to check if given info will lead to a valid configuration of @ref CLBatchToSpaceLayer * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -124,7 +135,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, int32_t block_shape_x, int32_t block_shape_y, const ITensorInfo *output, const CropInfo &crop_info = CropInfo{}); + static Status validate(const ITensorInfo *input, + int32_t block_shape_x, + int32_t block_shape_y, + const ITensorInfo *output, + const CropInfo &crop_info = CropInfo{}); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLBitwiseAnd.h b/arm_compute/runtime/CL/functions/CLBitwiseAnd.h index b30be9b24f..f82af3af9b 100644 --- a/arm_compute/runtime/CL/functions/CLBitwiseAnd.h +++ b/arm_compute/runtime/CL/functions/CLBitwiseAnd.h @@ -61,7 +61,10 @@ public: * @param[in] input2 Input tensor. Data types supported: U8. * @param[out] output Output tensor. Data types supported: U8. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLBITWISEAND_H */ diff --git a/arm_compute/runtime/CL/functions/CLBitwiseNot.h b/arm_compute/runtime/CL/functions/CLBitwiseNot.h index 1456ebe57e..31f8e86802 100644 --- a/arm_compute/runtime/CL/functions/CLBitwiseNot.h +++ b/arm_compute/runtime/CL/functions/CLBitwiseNot.h @@ -60,5 +60,5 @@ public: */ void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLBITWISENOT_H */ diff --git a/arm_compute/runtime/CL/functions/CLBitwiseOr.h b/arm_compute/runtime/CL/functions/CLBitwiseOr.h index ff0a1f0d73..9a25a2099e 100644 --- a/arm_compute/runtime/CL/functions/CLBitwiseOr.h +++ b/arm_compute/runtime/CL/functions/CLBitwiseOr.h @@ -61,7 +61,10 @@ public: * @param[in] input2 Input tensor. Data types supported: U8. * @param[out] output Output tensor. Data types supported: U8. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLBITWISEOR_H */ diff --git a/arm_compute/runtime/CL/functions/CLBitwiseXor.h b/arm_compute/runtime/CL/functions/CLBitwiseXor.h index 0cd9d073b4..9e288ef7b6 100644 --- a/arm_compute/runtime/CL/functions/CLBitwiseXor.h +++ b/arm_compute/runtime/CL/functions/CLBitwiseXor.h @@ -61,7 +61,10 @@ public: * @param[in] input2 Input tensor. Data types supported: U8. * @param[out] output Output tensor. Data types supported: U8. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLBITWISEXOR_H */ diff --git a/arm_compute/runtime/CL/functions/CLBoundingBoxTransform.h b/arm_compute/runtime/CL/functions/CLBoundingBoxTransform.h index d3499c3949..dba5497f5d 100644 --- a/arm_compute/runtime/CL/functions/CLBoundingBoxTransform.h +++ b/arm_compute/runtime/CL/functions/CLBoundingBoxTransform.h @@ -64,7 +64,10 @@ public: * * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the BoundingBoxTransformInfo struct. */ - void configure(const ICLTensor *boxes, ICLTensor *pred_boxes, const ICLTensor *deltas, const BoundingBoxTransformInfo &info); + void configure(const ICLTensor *boxes, + ICLTensor *pred_boxes, + const ICLTensor *deltas, + const BoundingBoxTransformInfo &info); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -76,7 +79,11 @@ public: * * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the BoundingBoxTransformInfo struct. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *boxes, ICLTensor *pred_boxes, const ICLTensor *deltas, const BoundingBoxTransformInfo &info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *boxes, + ICLTensor *pred_boxes, + const ICLTensor *deltas, + const BoundingBoxTransformInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLBoundingBoxTransform * @@ -90,7 +97,10 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *boxes, const ITensorInfo *pred_boxes, const ITensorInfo *deltas, const BoundingBoxTransformInfo &info); + static Status validate(const ITensorInfo *boxes, + const ITensorInfo *pred_boxes, + const ITensorInfo *deltas, + const BoundingBoxTransformInfo &info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLBOUNDINGBOXTRANSFORM_H */ diff --git a/arm_compute/runtime/CL/functions/CLCast.h b/arm_compute/runtime/CL/functions/CLCast.h index 650cd11b9b..9433f08fac 100644 --- a/arm_compute/runtime/CL/functions/CLCast.h +++ b/arm_compute/runtime/CL/functions/CLCast.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLCAST_H #define ARM_COMPUTE_CLCAST_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -79,7 +78,8 @@ public: */ void configure(const ICLTensor *input, ICLTensor *output, ConvertPolicy policy); // Initialize the function's source, destination - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, ConvertPolicy policy); + void + configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, ConvertPolicy policy); /** Static function to check if given info will lead to a valid configuration of @ref CLCast * * @param[in] input Source tensor info. Data types supported: U8/S8/U16/S16/U32/S32/U64/S64/F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLChannelShuffleLayer.h b/arm_compute/runtime/CL/functions/CLChannelShuffleLayer.h index 3dc62595d2..8ca848a020 100644 --- a/arm_compute/runtime/CL/functions/CLChannelShuffleLayer.h +++ b/arm_compute/runtime/CL/functions/CLChannelShuffleLayer.h @@ -65,7 +65,10 @@ public: * @param[out] output Output tensor. Data type supported: Same as @p input * @param[in] num_groups Number of groups. Must be greater than 1 and the number of channels of the tensors must be a multiple of the number of groups. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, unsigned int num_groups); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + unsigned int num_groups); /** Static function to check if given info will lead to a valid configuration of @ref CLChannelShuffleLayerKernel * * @param[in] input Input tensor info. Data types supported: All. diff --git a/arm_compute/runtime/CL/functions/CLComparison.h b/arm_compute/runtime/CL/functions/CLComparison.h index 3f984900ee..fca4b168b0 100644 --- a/arm_compute/runtime/CL/functions/CLComparison.h +++ b/arm_compute/runtime/CL/functions/CLComparison.h @@ -66,7 +66,11 @@ public: * @param[out] output Destination tensor. Data types supported: U8. * @param[out] operation Comparison operation to be used. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, ComparisonOperation operation); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + ComparisonOperation operation); /** Static function to check if given info will lead to a valid configuration of @ref CLComparison * * @param[in] input1 Source tensor. Data types supported: All. @@ -76,7 +80,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ComparisonOperation operation); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + ComparisonOperation operation); }; /** Basic function to run @ref CLComparisonKernel */ diff --git a/arm_compute/runtime/CL/functions/CLConcatenateLayer.h b/arm_compute/runtime/CL/functions/CLConcatenateLayer.h index 71e84e21b5..88c4bed595 100644 --- a/arm_compute/runtime/CL/functions/CLConcatenateLayer.h +++ b/arm_compute/runtime/CL/functions/CLConcatenateLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLCONCATENATELAYER_H #define ARM_COMPUTE_CLCONCATENATELAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> #include <vector> @@ -95,7 +94,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input. * @param[in] axis Concatenation axis. Supported underlying concatenation axis are 0, 1, 2 and 3. */ - void configure(const CLCompileContext &compile_context, std::vector<const ICLTensor *> &inputs_vector, ICLTensor *output, size_t axis); + void configure(const CLCompileContext &compile_context, + std::vector<const ICLTensor *> &inputs_vector, + ICLTensor *output, + size_t axis); /** Static function to check if given info will lead to a valid configuration of @ref CLConcatenateLayer * * @note Input and output tensor dimensions preconditions defer depending on the concatenation axis. @@ -108,7 +110,8 @@ public: * * @return a status */ - static Status validate(const std::vector<const ITensorInfo *> &inputs_vector, const ITensorInfo *output, size_t axis); + static Status + validate(const std::vector<const ITensorInfo *> &inputs_vector, const ITensorInfo *output, size_t axis); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLConv3D.h b/arm_compute/runtime/CL/functions/CLConv3D.h index 5728fe79d8..aabaf01ab7 100644 --- a/arm_compute/runtime/CL/functions/CLConv3D.h +++ b/arm_compute/runtime/CL/functions/CLConv3D.h @@ -77,20 +77,33 @@ public: * @param[in] conv3d_info Contains strides, padding, rounding, activation, dilation and fast math information. Activation and fast math are currently unused. * */ - void configure(const CLCompileContext &compile_context, const ICLTensor *src, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *dst, const Conv3dInfo &conv3d_info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *src, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *dst, + const Conv3dInfo &conv3d_info); /** Set the src and dst tensors. * * Similar to CLConv3D::configure() but using the default compile context * */ - void configure(const ICLTensor *src, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *dst, const Conv3dInfo &conv3d_info); + void configure(const ICLTensor *src, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *dst, + const Conv3dInfo &conv3d_info); /** Static function to check if given info will lead to a valid configuration of @ref CLConv3D * * Similar to CLConv3D::configure() * * @return a status */ - static Status validate(const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst, const Conv3dInfo &conv3d_info); + static Status validate(const ITensorInfo *src, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *dst, + const Conv3dInfo &conv3d_info); // Inherited methods overridden: void run() override; @@ -99,5 +112,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLCONVOLUTION3DLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h b/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h index 0a634b5482..409430d595 100644 --- a/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h +++ b/arm_compute/runtime/CL/functions/CLConvertFullyConnectedWeights.h @@ -69,7 +69,10 @@ public: * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer). * @param[in] data_layout The data layout the weights have been trained in. */ - void configure(const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape, DataLayout data_layout); + void configure(const ICLTensor *input, + ICLTensor *output, + const TensorShape &original_input_shape, + DataLayout data_layout); /** Initialize the function. * * @param[in] compile_context The compile context to be used. @@ -78,7 +81,11 @@ public: * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer). * @param[in] data_layout The data layout the weights have been trained in. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const TensorShape &original_input_shape, DataLayout data_layout); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const TensorShape &original_input_shape, + DataLayout data_layout); /** Static function to check if given info will lead to a valid configuration of @ref CLConvertFullyConnectedWeights * * @param[in] input Source weights tensor info to convert. Must be 2 dimensional. Data types supported: All. @@ -86,7 +93,10 @@ public: * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer). * @param[in] data_layout The data layout the weights have been trained in. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const TensorShape &original_input_shape, DataLayout data_layout); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const TensorShape &original_input_shape, + DataLayout data_layout); // Inherited methods overridden: void run() override; @@ -144,7 +154,10 @@ public: * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer). * @param[in] data_layout The data layout the weights have been trained in. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const TensorShape &original_input_shape, DataLayout data_layout) + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const TensorShape &original_input_shape, + DataLayout data_layout) { _func.configure(compile_context, input, &_output, original_input_shape, data_layout); } diff --git a/arm_compute/runtime/CL/functions/CLConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLConvolutionLayer.h index 77bf48d613..8487be71c3 100644 --- a/arm_compute/runtime/CL/functions/CLConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLConvolutionLayer.h @@ -120,8 +120,16 @@ public: * available which may introduce a drop of accuracy as well. Default is false * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, const WeightsInfo &weights_info = WeightsInfo(), - const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, unsigned int num_groups = 1); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -142,9 +150,17 @@ public: * available which may introduce a drop of accuracy as well. Default is false * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, - unsigned int num_groups = 1); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if given info will lead to a valid configuration of @ref CLConvolutionLayer * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], @@ -166,9 +182,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, - unsigned int num_groups = 1); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if given info will return the convolution called by @ref CLConvolutionLayer * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], @@ -188,8 +211,15 @@ public: * * @return the Convolution Method Hint */ - static ConvolutionMethod get_convolution_method(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info, const ActivationLayerInfo &act_info, const GPUTarget gpu_target, const Size2D &dilation = Size2D(1U, 1U), bool enable_fast_math = false); + static ConvolutionMethod get_convolution_method(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info, + const ActivationLayerInfo &act_info, + const GPUTarget gpu_target, + const Size2D &dilation = Size2D(1U, 1U), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; void prepare() override; diff --git a/arm_compute/runtime/CL/functions/CLCopy.h b/arm_compute/runtime/CL/functions/CLCopy.h index 4fc4183d3e..fd40b7b9de 100644 --- a/arm_compute/runtime/CL/functions/CLCopy.h +++ b/arm_compute/runtime/CL/functions/CLCopy.h @@ -27,6 +27,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/core/Window.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -74,7 +75,10 @@ public: * @param[in] dst_window (Optional) Window to be used in case only copying into part of a tensor. Default is nullptr. * */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, Window *dst_window = nullptr); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + Window *dst_window = nullptr); /** Static function to check if given info will lead to a valid configuration of @ref CLCopy * * @param[in] input Source tensor. Data types supported: All. diff --git a/arm_compute/runtime/CL/functions/CLCrop.h b/arm_compute/runtime/CL/functions/CLCrop.h index a474215190..2942e9362a 100644 --- a/arm_compute/runtime/CL/functions/CLCrop.h +++ b/arm_compute/runtime/CL/functions/CLCrop.h @@ -27,6 +27,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/core/Window.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -71,7 +72,13 @@ public: * @param[in] extrapolation_value Value to be used for values outside of the image. Default is 0. * @param[in] output_window Output window to be used in case cropped image is being copied into a tensor. Default is nullptr. */ - void configure(const ICLTensor *input, ICLTensor *output, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value = 0, Window *output_window = nullptr); + void configure(const ICLTensor *input, + ICLTensor *output, + Coordinates2D start, + Coordinates2D end, + uint32_t batch_index, + float extrapolation_value = 0, + Window *output_window = nullptr); /** Configure function * * @note Supported tensor rank: up to 4 @@ -85,8 +92,14 @@ public: * @param[in] extrapolation_value Value to be used for values outside of the image. Default is 0. * @param[in] output_window Output window to be used in case cropped image is being copied into a tensor. Default is nullptr. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value = 0, - Window *output_window = nullptr); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + Coordinates2D start, + Coordinates2D end, + uint32_t batch_index, + float extrapolation_value = 0, + Window *output_window = nullptr); /** Static function to check if given info will lead to a valid configuration of @ref CLStridedSliceKernel * @@ -100,8 +113,13 @@ public: * @param[in] extrapolation_value Value to be used for values outside of the image. Default is 0. * @param[in] output_window Output window to be used in case cropped image is being copied into a tensor. Default is nullptr. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, Coordinates2D start, Coordinates2D end, uint32_t batch_index, float extrapolation_value = 0, - Window *output_window = nullptr); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + Coordinates2D start, + Coordinates2D end, + uint32_t batch_index, + float extrapolation_value = 0, + Window *output_window = nullptr); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLCropResize.h b/arm_compute/runtime/CL/functions/CLCropResize.h index 5c60c2879c..6fb055e893 100644 --- a/arm_compute/runtime/CL/functions/CLCropResize.h +++ b/arm_compute/runtime/CL/functions/CLCropResize.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_CL_CROP_RESIZE_H #include "arm_compute/core/CL/ICLTensor.h" - #include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/CL/functions/CLCopy.h" #include "arm_compute/runtime/CL/functions/CLCrop.h" @@ -82,8 +81,13 @@ public: * @param[in] method The policy to be used when resizing image. Default is bilinear. * @param[in] extrapolation_value Value to be used for values outside of the image for cropping and resizing. Default is 0. */ - void configure(const ICLTensor *input, ICLTensor *boxes, ICLTensor *box_ind, ICLTensor *output, Coordinates2D crop_size, - InterpolationPolicy method = InterpolationPolicy::BILINEAR, float extrapolation_value = 0); + void configure(const ICLTensor *input, + ICLTensor *boxes, + ICLTensor *box_ind, + ICLTensor *output, + Coordinates2D crop_size, + InterpolationPolicy method = InterpolationPolicy::BILINEAR, + float extrapolation_value = 0); /** Configure kernel * * @note Supported tensor rank: up to 4 @@ -100,8 +104,14 @@ public: * @param[in] method The policy to be used when resizing image. Default is bilinear. * @param[in] extrapolation_value Value to be used for values outside of the image for cropping and resizing. Default is 0. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *boxes, ICLTensor *box_ind, ICLTensor *output, Coordinates2D crop_size, - InterpolationPolicy method = InterpolationPolicy::BILINEAR, float extrapolation_value = 0); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *boxes, + ICLTensor *box_ind, + ICLTensor *output, + Coordinates2D crop_size, + InterpolationPolicy method = InterpolationPolicy::BILINEAR, + float extrapolation_value = 0); /** Static function to check if given info will lead to a valid configuration of @ref NESlice * @@ -121,8 +131,13 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, ITensorInfo *boxes, ITensorInfo *box_ind, const ITensorInfo *output, - Coordinates2D crop_size, InterpolationPolicy method, float extrapolation_value); + static Status validate(const ITensorInfo *input, + ITensorInfo *boxes, + ITensorInfo *box_ind, + const ITensorInfo *output, + Coordinates2D crop_size, + InterpolationPolicy method, + float extrapolation_value); void run() override; diff --git a/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h b/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h index 0c59e2c86d..92f87ee461 100644 --- a/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLDeconvolutionLayer.h @@ -70,7 +70,12 @@ public: * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref opencl::kernels::ClWeightsReshapeKernel. * */ - void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info, const WeightsInfo &weights_info = WeightsInfo()); + void configure(ICLTensor *input, + ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &deconv_info, + const WeightsInfo &weights_info = WeightsInfo()); /** Set the input, weights, biases and output tensors. * * @param[in] compile_context The compile context to be used. @@ -82,8 +87,13 @@ public: * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref opencl::kernels::ClWeightsReshapeKernel. * */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info, - const WeightsInfo &weights_info = WeightsInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &deconv_info, + const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. Data types supported: QASYMM8_SIGNED/QASYMM8/F16/F32. @@ -95,11 +105,19 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &deconv_info, - const WeightsInfo &weights_info = WeightsInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + ITensorInfo *output, + const PadStrideInfo &deconv_info, + const WeightsInfo &weights_info = WeightsInfo()); - static DeconvolutionMethod get_deconvolution_method(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &deconv_info, - const WeightsInfo &weights_info); + static DeconvolutionMethod get_deconvolution_method(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + ITensorInfo *output, + const PadStrideInfo &deconv_info, + const WeightsInfo &weights_info); // Inherited methods overridden: void run() override; void prepare() override; diff --git a/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h b/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h index 344ebd0afb..5a2abafe79 100644 --- a/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h +++ b/arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h @@ -82,7 +82,8 @@ public: * @param[out] output Destination tensor. Data type supported: same as @p input. * @param[in] info Contains padding and policies to be used in the deconvolution. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const PadStrideInfo &info); + void + configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const PadStrideInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayerUpsample * * @param[in] input Source tensor info. Data type supported: All. diff --git a/arm_compute/runtime/CL/functions/CLDepthConvertLayer.h b/arm_compute/runtime/CL/functions/CLDepthConvertLayer.h index 58deb7ec40..3e7ca8830b 100644 --- a/arm_compute/runtime/CL/functions/CLDepthConvertLayer.h +++ b/arm_compute/runtime/CL/functions/CLDepthConvertLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLDEPTHCONVERT_H #define ARM_COMPUTE_CLDEPTHCONVERT_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -96,7 +95,11 @@ public: * @param[in] policy Conversion policy. * @param[in] shift Value for down/up conversions. Must be 0 <= shift < 8. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, ConvertPolicy policy, uint32_t shift); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + ConvertPolicy policy, + uint32_t shift); /** Static function to check if given info will lead to a valid configuration of @ref CLDepthConvertLayer * * @param[in] input Source tensor info. Data types supported: U8/S8/U16/S16/U32/S32/F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h b/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h index 0026cc2b67..14d0a7ec7c 100644 --- a/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h +++ b/arm_compute/runtime/CL/functions/CLDepthToSpaceLayer.h @@ -60,7 +60,8 @@ public: * @param[out] output Tensor output. Data types supported: same as @p input * @param[in] block_shape Block shape value. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t block_shape); + void + configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t block_shape); /** Static function to check if given info will lead to a valid configuration of @ref CLDepthToSpaceLayer. * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -71,5 +72,5 @@ public: */ static Status validate(const ITensorInfo *input, const ITensorInfo *output, int32_t block_shape); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLDEPTHTOSPACELAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLDepthwiseConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLDepthwiseConvolutionLayer.h index 2798449100..2c0fa7aa22 100644 --- a/arm_compute/runtime/CL/functions/CLDepthwiseConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLDepthwiseConvolutionLayer.h @@ -88,15 +88,28 @@ public: * * @note: For in-place support, please check @ref CLDepthwiseConvolutionLayerNativeKernel */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + ActivationLayerInfo act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); /** Initialize the function's source, destination, weights and convolution information. * * Similar to @ref CLDepthwiseConvolutionLayer::configure() */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + ActivationLayerInfo act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); /** Static function to check if given info will lead to a valid configuration of @ref CLDepthwiseConvolutionLayer * @@ -104,8 +117,14 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, ActivationLayerInfo act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + ActivationLayerInfo act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); // Inherited methods overriden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLDirectConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLDirectConvolutionLayer.h index 462a3ac07e..84900b03a3 100644 --- a/arm_compute/runtime/CL/functions/CLDirectConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLDirectConvolutionLayer.h @@ -79,7 +79,12 @@ public: * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -94,7 +99,12 @@ public: * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLDirectConvolutionLayer * @@ -111,7 +121,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: @@ -121,5 +135,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLDIRECTCONVOLUTIONLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLDirectDeconvolutionLayer.h b/arm_compute/runtime/CL/functions/CLDirectDeconvolutionLayer.h index d0a61cdd36..14384a09b5 100644 --- a/arm_compute/runtime/CL/functions/CLDirectDeconvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLDirectDeconvolutionLayer.h @@ -24,12 +24,11 @@ #ifndef ARM_COMPUTE_CLDIRECTDECONVOLUTIONLAYER_H #define ARM_COMPUTE_CLDIRECTDECONVOLUTIONLAYER_H +#include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/CL/functions/CLConvolutionLayer.h" #include "arm_compute/runtime/CL/functions/CLDeconvolutionLayerUpsample.h" #include "arm_compute/runtime/CL/functions/CLReverse.h" #include "arm_compute/runtime/CL/functions/CLTranspose.h" - -#include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" @@ -111,7 +110,12 @@ public: * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref opencl::kernels::ClWeightsReshapeKernel. * */ - void configure(ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &info, const WeightsInfo &weights_info = WeightsInfo()); + void configure(ICLTensor *input, + ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &info, + const WeightsInfo &weights_info = WeightsInfo()); /** Set the input, weights, biases and output tensors. * * @param[in] compile_context The compile context to be used. @@ -125,8 +129,13 @@ public: * @param[in] weights_info (Optional) Weights information needed for @ref CLConvolutionLayer, specifies if the weights tensor has been reshaped with @ref opencl::kernels::ClWeightsReshapeKernel. * */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &info, - const WeightsInfo &weights_info = WeightsInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &info, + const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLDirectDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. @@ -140,8 +149,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, ITensorInfo *output, const PadStrideInfo &info, - const WeightsInfo &weights_info = WeightsInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + ITensorInfo *output, + const PadStrideInfo &info, + const WeightsInfo &weights_info = WeightsInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLElementwiseOperations.h b/arm_compute/runtime/CL/functions/CLElementwiseOperations.h index 9de362d2b2..13844c98a1 100644 --- a/arm_compute/runtime/CL/functions/CLElementwiseOperations.h +++ b/arm_compute/runtime/CL/functions/CLElementwiseOperations.h @@ -82,7 +82,11 @@ public: * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * Valid configurations (Input1,Input2) -> Output : @@ -108,7 +112,11 @@ public: * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, ConvertPolicy policy, + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output, + ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClSaturatedArithmeticKernel for addition * @@ -134,7 +142,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -192,7 +204,11 @@ public: * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * Valid configurations (Input1,Input2) -> Output : @@ -218,7 +234,11 @@ public: * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, ConvertPolicy policy, + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output, + ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClSaturatedArithmeticKernel for subtraction * @@ -244,7 +264,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -292,7 +316,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output. * * @param[in] compile_context The compile context to be used. @@ -303,7 +330,11 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLArithmeticDivision * * @param[in] input1 First tensor input info. Data types supported: F16/F32. @@ -313,7 +344,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -368,7 +402,10 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * @param[in] compile_context The compile context to be used. @@ -379,7 +416,11 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClArithmeticKernel for max * * @param[in] input1 First tensor input info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/U32/F16/F32. @@ -389,7 +430,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -444,7 +488,10 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * @param[in] compile_context The compile context to be used. @@ -455,7 +502,11 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClArithmeticKernel for min * * @param[in] input1 First tensor input info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/S32/U32/F16/F32. @@ -465,7 +516,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -518,7 +572,10 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * @param[in] compile_context The compile context to be used. @@ -529,7 +586,11 @@ public: * @param[out] output Output tensor. Data types supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClArithmeticKernel for squared difference * * @param[in] input1 First tensor input info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32. @@ -539,7 +600,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -587,7 +651,10 @@ public: * @param[out] output Output tensor. Data types supported:F16/F32. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and conversion policy. * * @param[in] compile_context The compile context to be used. @@ -598,7 +665,11 @@ public: * @param[out] output Output tensor. Data types supported:F16/F32. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClArithmeticKernel for power * * @param[in] input1 First tensor input info. Data types supported: F16/F32. @@ -608,7 +679,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLElementwiseUnaryLayer.h b/arm_compute/runtime/CL/functions/CLElementwiseUnaryLayer.h index 594ee4cfdc..d186b70d93 100644 --- a/arm_compute/runtime/CL/functions/CLElementwiseUnaryLayer.h +++ b/arm_compute/runtime/CL/functions/CLElementwiseUnaryLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLELEMENTWISEUNARYLAYER_H #define ARM_COMPUTE_CLELEMENTWISEUNARYLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/CL/functions/CLFFT1D.h b/arm_compute/runtime/CL/functions/CLFFT1D.h index c7112dc737..49ecf3c260 100644 --- a/arm_compute/runtime/CL/functions/CLFFT1D.h +++ b/arm_compute/runtime/CL/functions/CLFFT1D.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_CLFFT1D_H #define ARM_COMPUTE_CLFFT1D_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/FunctionDescriptors.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" namespace arm_compute @@ -82,7 +81,10 @@ public: * @param[out] output Destination tensor. Data types and data layouts supported: Same as @p input. * @param[in] config FFT related configuration */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const FFT1DInfo &config); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const FFT1DInfo &config); /** Static function to check if given info will lead to a valid configuration of @ref CLFFT1D. * * @param[in] input Source tensor info. Data types supported: F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLFFT2D.h b/arm_compute/runtime/CL/functions/CLFFT2D.h index 3d20327bf1..b7d15f1602 100644 --- a/arm_compute/runtime/CL/functions/CLFFT2D.h +++ b/arm_compute/runtime/CL/functions/CLFFT2D.h @@ -24,11 +24,10 @@ #ifndef ARM_COMPUTE_CLFFT2D_H #define ARM_COMPUTE_CLFFT2D_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/CL/functions/CLFFT1D.h" #include "arm_compute/runtime/FunctionDescriptors.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" namespace arm_compute @@ -79,7 +78,10 @@ public: * @param[out] output Destination tensor. Data types and data layouts supported: Same as @p input. * @param[in] config FFT related configuration */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const FFT2DInfo &config); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const FFT2DInfo &config); /** Static function to check if given info will lead to a valid configuration of @ref CLFFT2D. * * @param[in] input Source tensor info. Data types supported: F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h index f873cb0b86..ed78bbb7a7 100644 --- a/arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLFFTConvolutionLayer.h @@ -24,8 +24,6 @@ #ifndef ARM_COMPUTE_CLFFTCONVOLUTIONLAYER_H #define ARM_COMPUTE_CLFFTCONVOLUTIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/runtime/CL/functions/CLActivationLayer.h" #include "arm_compute/runtime/CL/functions/CLElementwiseOperations.h" @@ -37,6 +35,7 @@ #include "arm_compute/runtime/CL/functions/CLReshapeLayer.h" #include "arm_compute/runtime/CL/functions/CLReverse.h" #include "arm_compute/runtime/CL/functions/CLSlice.h" +#include "arm_compute/runtime/IFunction.h" namespace arm_compute { @@ -94,8 +93,13 @@ public: * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation * available which may introduce a drop of accuracy as well. Default is false */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); /** Set the input and output tensors. * * @note: This function only works with any square kernel size and unit strides for both NCHW and NHWC data layout @@ -113,8 +117,14 @@ public: * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation * available which may introduce a drop of accuracy as well. Default is false */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); /** Static function to check if given info will lead to a valid configuration of @ref CLFFTConvolutionLayer * * @note: This function only works with any square kernel size and unit strides for both NCHW and NHWC data layout @@ -133,8 +143,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLFill.h b/arm_compute/runtime/CL/functions/CLFill.h index 341d93a9f6..be1059761a 100644 --- a/arm_compute/runtime/CL/functions/CLFill.h +++ b/arm_compute/runtime/CL/functions/CLFill.h @@ -28,6 +28,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/core/Window.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -73,7 +74,10 @@ public: * @param[in] constant_value The value used to fill the planes of the tensor * @param[in] window Window to be used in case setting only part of a tensor. Default is nullptr. */ - void configure(const CLCompileContext &compile_context, ICLTensor *tensor, const PixelValue &constant_value, Window *window = nullptr); + void configure(const CLCompileContext &compile_context, + ICLTensor *tensor, + const PixelValue &constant_value, + Window *window = nullptr); /** Static function to check if given info will lead to a valid configuration of @ref CLFill * * @param[in] tensor Source tensor info. Data types supported: All. diff --git a/arm_compute/runtime/CL/functions/CLFloor.h b/arm_compute/runtime/CL/functions/CLFloor.h index 87cd5b44c7..4d3d704857 100644 --- a/arm_compute/runtime/CL/functions/CLFloor.h +++ b/arm_compute/runtime/CL/functions/CLFloor.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLFLOOR_H #define ARM_COMPUTE_CLFLOOR_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/CL/functions/CLFullyConnectedLayer.h b/arm_compute/runtime/CL/functions/CLFullyConnectedLayer.h index b784226a2f..9fd0b4aaef 100644 --- a/arm_compute/runtime/CL/functions/CLFullyConnectedLayer.h +++ b/arm_compute/runtime/CL/functions/CLFullyConnectedLayer.h @@ -25,9 +25,8 @@ #define ARM_COMPUTE_CLFULLYCONNECTEDLAYER_H #include "arm_compute/function_info/FullyConnectedLayerInfo.h" -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/CL/CLTensor.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IWeightsManager.h" #include "arm_compute/runtime/MemoryGroup.h" @@ -45,7 +44,8 @@ class CLFullyConnectedLayer : public IFunction { public: /** Constructor */ - CLFullyConnectedLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, IWeightsManager *weights_manager = nullptr); + CLFullyConnectedLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, + IWeightsManager *weights_manager = nullptr); /** Default destructor */ ~CLFullyConnectedLayer(); /** Prevent instances of this class from being copied (As this class contains pointers) */ @@ -83,13 +83,20 @@ public: * Data type supported: Same as @p input. * @param[in] fc_info (Optional) Fully connected layer additional info */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo()); /** Set the input and output tensors. * * Similar to @ref CLFullyConnectedLayer */ - void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, + void configure(const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLFullyConnectedLayer * @@ -97,7 +104,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo()); //Inherited methods override diff --git a/arm_compute/runtime/CL/functions/CLFuseBatchNormalization.h b/arm_compute/runtime/CL/functions/CLFuseBatchNormalization.h index cd75270392..2e777273cd 100644 --- a/arm_compute/runtime/CL/functions/CLFuseBatchNormalization.h +++ b/arm_compute/runtime/CL/functions/CLFuseBatchNormalization.h @@ -78,9 +78,16 @@ public: * @param[in] epsilon (Optional) Batch normalization layer epsilon parameter. Defaults to 0.001f. * @param[in] fbn_type (Optional) Fused batch normalization type. Defaults to Convolution. */ - void configure(const ICLTensor *input_weights, const ICLTensor *bn_mean, const ICLTensor *bn_var, ICLTensor *fused_weights, ICLTensor *fused_bias, - const ICLTensor *input_bias = nullptr, const ICLTensor *bn_beta = nullptr, const ICLTensor *bn_gamma = nullptr, - float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); + void configure(const ICLTensor *input_weights, + const ICLTensor *bn_mean, + const ICLTensor *bn_var, + ICLTensor *fused_weights, + ICLTensor *fused_bias, + const ICLTensor *input_bias = nullptr, + const ICLTensor *bn_beta = nullptr, + const ICLTensor *bn_gamma = nullptr, + float epsilon = 0.001f, + FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -97,9 +104,17 @@ public: * @param[in] epsilon (Optional) Batch normalization layer epsilon parameter. Defaults to 0.001f. * @param[in] fbn_type (Optional) Fused batch normalization type. Defaults to Convolution. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input_weights, const ICLTensor *bn_mean, const ICLTensor *bn_var, ICLTensor *fused_weights, ICLTensor *fused_bias, - const ICLTensor *input_bias = nullptr, const ICLTensor *bn_beta = nullptr, const ICLTensor *bn_gamma = nullptr, - float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input_weights, + const ICLTensor *bn_mean, + const ICLTensor *bn_var, + ICLTensor *fused_weights, + ICLTensor *fused_bias, + const ICLTensor *input_bias = nullptr, + const ICLTensor *bn_beta = nullptr, + const ICLTensor *bn_gamma = nullptr, + float epsilon = 0.001f, + FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); /** Static function to check if given info will lead to a valid configuration of @ref CLFuseBatchNormalization * * @param[in] input_weights Input weights tensor info for convolution or depthwise convolution layer. Data type supported: F16/F32. Data layout supported: NCHW, NHWC @@ -117,10 +132,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input_weights, const ITensorInfo *bn_mean, const ITensorInfo *bn_var, - const ITensorInfo *fused_weights, const ITensorInfo *fused_bias, - const ITensorInfo *input_bias = nullptr, const ITensorInfo *bn_beta = nullptr, const ITensorInfo *bn_gamma = nullptr, - float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); + static Status validate(const ITensorInfo *input_weights, + const ITensorInfo *bn_mean, + const ITensorInfo *bn_var, + const ITensorInfo *fused_weights, + const ITensorInfo *fused_bias, + const ITensorInfo *input_bias = nullptr, + const ITensorInfo *bn_beta = nullptr, + const ITensorInfo *bn_gamma = nullptr, + float epsilon = 0.001f, + FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGEMM.h b/arm_compute/runtime/CL/functions/CLGEMM.h index 3a39aca692..f5e6aa1237 100644 --- a/arm_compute/runtime/CL/functions/CLGEMM.h +++ b/arm_compute/runtime/CL/functions/CLGEMM.h @@ -92,13 +92,26 @@ public: * if the reshape of matrix B should happen only for the first run. GEMMInfo also contains information about the reshaping * in case matrix A and matrix B have been already transformed. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *a, const ICLTensor *b, const ICLTensor *c, ICLTensor *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + void configure(const CLCompileContext &compile_context, + const ICLTensor *a, + const ICLTensor *b, + const ICLTensor *c, + ICLTensor *output, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); /** Initialise the kernel's inputs and output * * Similar to @ref CLGEMM::configure() */ - void configure(const ICLTensor *a, const ICLTensor *b, const ICLTensor *c, ICLTensor *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + void configure(const ICLTensor *a, + const ICLTensor *b, + const ICLTensor *c, + ICLTensor *output, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLGEMM. * @@ -106,7 +119,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + static Status validate(const ITensorInfo *a, + const ITensorInfo *b, + const ITensorInfo *c, + const ITensorInfo *output, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h index 4bafef27a9..70ceb1513b 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h @@ -52,7 +52,8 @@ public: * @param[in] memory_manager (Optional) Memory manager. * @param[in] weights_manager (Optional) Weights manager. */ - CLGEMMConvolutionLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, IWeightsManager *weights_manager = nullptr); + CLGEMMConvolutionLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, + IWeightsManager *weights_manager = nullptr); /** Prevent instances of this class from being copied (As this class contains pointers) */ CLGEMMConvolutionLayer(const CLGEMMConvolutionLayer &) = delete; /** Default move constructor */ @@ -95,8 +96,15 @@ public: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout */ - void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, const WeightsInfo &weights_info = WeightsInfo(), - const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), unsigned int num_groups = 1); + void configure(const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + unsigned int num_groups = 1); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -116,9 +124,16 @@ public: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is only supported for NCHW data layout */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), - const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), unsigned int num_groups = 1); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + unsigned int num_groups = 1); /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMConvolutionLayer. * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], @@ -139,8 +154,15 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), unsigned int num_groups = 1); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + unsigned int num_groups = 1); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h index c985738a9c..3e8929c5ad 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h @@ -113,7 +113,11 @@ public: * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. Data layout supported: same as @p input. * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. This function supports only stride_x = weights.width && stride_y = weights.height. Moreover, padding is not supported. */ - void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info); + void configure(const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &deconv_info); /** Set the input, weights, biases and output tensors. * * @param[in] compile_context The compile context to be used. @@ -124,7 +128,12 @@ public: * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. Data layout supported: same as @p input. * @param[in] deconv_info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. This function supports only stride_x = weights.width && stride_y = weights.height. Moreover, padding is not supported. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *bias, + ICLTensor *output, + const PadStrideInfo &deconv_info); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. @@ -136,7 +145,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, const ITensorInfo *output, const PadStrideInfo &deconv_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + const ITensorInfo *output, + const PadStrideInfo &deconv_info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h index 8b8d9f235f..1b8e5dcc1d 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h +++ b/arm_compute/runtime/CL/functions/CLGEMMLowpMatrixMultiplyCore.h @@ -91,7 +91,11 @@ public: * @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and * if the reshape of matrix B should be executed only for the first run */ - void configure(const ICLTensor *a, const ICLTensor *b, const ICLTensor *c, ICLTensor *output, const GEMMInfo &gemm_info = GEMMInfo()); + void configure(const ICLTensor *a, + const ICLTensor *b, + const ICLTensor *c, + ICLTensor *output, + const GEMMInfo &gemm_info = GEMMInfo()); /** Initialise the kernel's inputs, output * * @note GEMMLowp: low precision GEMM kernel. [A * B + C] @@ -110,7 +114,12 @@ public: * @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and * if the reshape of matrix B should be executed only for the first run */ - void configure(const CLCompileContext &compile_context, const ICLTensor *a, const ICLTensor *b, const ICLTensor *c, ICLTensor *output, const GEMMInfo &gemm_info = GEMMInfo()); + void configure(const CLCompileContext &compile_context, + const ICLTensor *a, + const ICLTensor *b, + const ICLTensor *c, + ICLTensor *output, + const GEMMInfo &gemm_info = GEMMInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMLowpMatrixMultiplyCore * * @param[in] a First input tensor info (Matrix A). Data type supported: QASYMM8. @@ -122,7 +131,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, const GEMMInfo &gemm_info = GEMMInfo()); + static Status validate(const ITensorInfo *a, + const ITensorInfo *b, + const ITensorInfo *c, + const ITensorInfo *output, + const GEMMInfo &gemm_info = GEMMInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h b/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h index 6ec7b71f7d..ff9c872896 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h +++ b/arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h @@ -85,7 +85,8 @@ public: * @param[out] output Output tensor. Data type supported: QASYMM8/QASYMM8_SIGNED/QSYMM16 * @param[in] info GEMMLowp output stage metadata. */ - void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const GEMMLowpOutputStageInfo &info); + void + configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const GEMMLowpOutputStageInfo &info); /** Initialise the kernel's inputs, output * * @param[in] compile_context The compile context to be used. @@ -95,7 +96,11 @@ public: * @param[out] output Output tensor. Data type supported: QASYMM8/QASYMM8_SIGNED * @param[in] info GEMMLowp output stage metadata. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, const GEMMLowpOutputStageInfo &info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *bias, + ICLTensor *output, + const GEMMLowpOutputStageInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref opencl::kernels::ClGemmLowpQuantizeDownInt32ScaleByFixedPointKernel * * @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32 @@ -106,7 +111,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo &info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *bias, + const ITensorInfo *output, + const GEMMLowpOutputStageInfo &info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLGather.h b/arm_compute/runtime/CL/functions/CLGather.h index 0f1ccbad08..360c8757b6 100644 --- a/arm_compute/runtime/CL/functions/CLGather.h +++ b/arm_compute/runtime/CL/functions/CLGather.h @@ -62,7 +62,11 @@ public: * @param[out] output Destination tensor. Data type supported: Same as @p input * @param[in] axis (Optional) The axis in @p input to gather @p indices from. Defaults to 0 */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *indices, ICLTensor *output, int axis = 0); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *indices, + ICLTensor *output, + int axis = 0); /** Static function to check if given info will lead to a valid configuration of @ref CLGatherKernel * @@ -73,7 +77,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *indices, const ITensorInfo *output, int axis = 0); + static Status + validate(const ITensorInfo *input, const ITensorInfo *indices, const ITensorInfo *output, int axis = 0); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLGATHER_H */ diff --git a/arm_compute/runtime/CL/functions/CLGenerateProposalsLayer.h b/arm_compute/runtime/CL/functions/CLGenerateProposalsLayer.h index aec5cdf1a8..3a201e79b0 100644 --- a/arm_compute/runtime/CL/functions/CLGenerateProposalsLayer.h +++ b/arm_compute/runtime/CL/functions/CLGenerateProposalsLayer.h @@ -100,7 +100,12 @@ public: * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the @ref GenerateProposalsInfo struct. * @note Proposals contains all the proposals. Of those, only the first num_valid_proposals are valid. */ - void configure(const ICLTensor *scores, const ICLTensor *deltas, const ICLTensor *anchors, ICLTensor *proposals, ICLTensor *scores_out, ICLTensor *num_valid_proposals, + void configure(const ICLTensor *scores, + const ICLTensor *deltas, + const ICLTensor *anchors, + ICLTensor *proposals, + ICLTensor *scores_out, + ICLTensor *num_valid_proposals, const GenerateProposalsInfo &info); /** Set the input and output tensors. * @@ -118,8 +123,14 @@ public: * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the @ref GenerateProposalsInfo struct. * @note Proposals contains all the proposals. Of those, only the first num_valid_proposals are valid. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *scores, const ICLTensor *deltas, const ICLTensor *anchors, ICLTensor *proposals, ICLTensor *scores_out, - ICLTensor *num_valid_proposals, const GenerateProposalsInfo &info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *scores, + const ICLTensor *deltas, + const ICLTensor *anchors, + ICLTensor *proposals, + ICLTensor *scores_out, + ICLTensor *num_valid_proposals, + const GenerateProposalsInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLGenerateProposalsLayer * @@ -135,7 +146,11 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *scores, const ITensorInfo *deltas, const ITensorInfo *anchors, const ITensorInfo *proposals, const ITensorInfo *scores_out, + static Status validate(const ITensorInfo *scores, + const ITensorInfo *deltas, + const ITensorInfo *anchors, + const ITensorInfo *proposals, + const ITensorInfo *scores_out, const ITensorInfo *num_valid_proposals, const GenerateProposalsInfo &info); diff --git a/arm_compute/runtime/CL/functions/CLIndirectConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLIndirectConvolutionLayer.h index 12b83ea25b..91952af5dc 100644 --- a/arm_compute/runtime/CL/functions/CLIndirectConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLIndirectConvolutionLayer.h @@ -75,7 +75,12 @@ public: * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -90,7 +95,12 @@ public: * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLIndirectConvolutionLayer * @@ -107,7 +117,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: @@ -117,5 +131,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLINDIRECTCONVOLUTIONLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLInstanceNormalizationLayer.h b/arm_compute/runtime/CL/functions/CLInstanceNormalizationLayer.h index 985a6a75f7..98d215dd4b 100644 --- a/arm_compute/runtime/CL/functions/CLInstanceNormalizationLayer.h +++ b/arm_compute/runtime/CL/functions/CLInstanceNormalizationLayer.h @@ -83,7 +83,12 @@ public: * @param[in] epsilon (Optional) Lower bound value for the normalization. Defaults to 1e-12 * @param[in] use_mixed_precision (Optional) Use mixed precision in case of FP16 execution */ - void configure(ICLTensor *input, ICLTensor *output, float gamma = 1.0f, float beta = 0.0f, float epsilon = 1e-12f, bool use_mixed_precision = true); + void configure(ICLTensor *input, + ICLTensor *output, + float gamma = 1.0f, + float beta = 0.0f, + float epsilon = 1e-12f, + bool use_mixed_precision = true); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -95,7 +100,13 @@ public: * @param[in] epsilon (Optional) Lower bound value for the normalization. Defaults to 1e-12 * @param[in] use_mixed_precision (Optional) Use mixed precision in case of FP16 execution */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, float gamma = 1.0f, float beta = 0.0f, float epsilon = 1e-12f, bool use_mixed_precision = true); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + float gamma = 1.0f, + float beta = 0.0f, + float epsilon = 1e-12f, + bool use_mixed_precision = true); /** Static function to check if given info will lead to a valid configuration of @ref CLInstanceNormalizationLayer. * @@ -108,8 +119,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, float gamma = 1.0f, float beta = 0.0f, float epsilon = 1e-12f, bool use_mixed_precision = true); - void run() override; + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + float gamma = 1.0f, + float beta = 0.0f, + float epsilon = 1e-12f, + bool use_mixed_precision = true); + void run() override; private: std::unique_ptr<ICLKernel> _inst_norm_kernel; /**< Kernel to run */ diff --git a/arm_compute/runtime/CL/functions/CLL2NormalizeLayer.h b/arm_compute/runtime/CL/functions/CLL2NormalizeLayer.h index 4dc5c778d2..a8b356a708 100644 --- a/arm_compute/runtime/CL/functions/CLL2NormalizeLayer.h +++ b/arm_compute/runtime/CL/functions/CLL2NormalizeLayer.h @@ -26,8 +26,8 @@ #include "arm_compute/core/Types.h" #include "arm_compute/runtime/CL/CLTensor.h" -#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include "arm_compute/runtime/CL/functions/CLReductionOperation.h" +#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" @@ -89,7 +89,8 @@ public: * @param[in] axis Axis along which to reduce. Negative values wrap around. Maximum supported actual reduction axis : 2 * @param[in] epsilon (Optional) Lower bound value for the normalization. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, int axis, float epsilon = 1e-12f); + void configure( + const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, int axis, float epsilon = 1e-12f); /** Static function to check if given info will lead to a valid configuration of @ref CLL2NormalizeLayer. * @@ -111,5 +112,5 @@ private: std::unique_ptr<CLL2NormalizeLayerKernel> _normalize_kernel; CLTensor _sumsq; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_CLL2NORMALIZELAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLLSTMLayer.h b/arm_compute/runtime/CL/functions/CLLSTMLayer.h index d26b4c5595..fe494991af 100644 --- a/arm_compute/runtime/CL/functions/CLLSTMLayer.h +++ b/arm_compute/runtime/CL/functions/CLLSTMLayer.h @@ -24,8 +24,6 @@ #ifndef ARM_COMPUTE_CLLSTMLAYER_H #define ARM_COMPUTE_CLLSTMLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/runtime/CL/CLTensor.h" #include "arm_compute/runtime/CL/functions/CLActivationLayer.h" @@ -37,9 +35,10 @@ #include "arm_compute/runtime/CL/functions/CLGEMM.h" #include "arm_compute/runtime/CL/functions/CLMeanStdDevNormalizationLayer.h" #include "arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h" +#include "arm_compute/runtime/common/LSTMParams.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" -#include "arm_compute/runtime/common/LSTMParams.h" #include <memory> @@ -53,7 +52,7 @@ namespace kernels { class ClTransposeKernel; } -} +} // namespace opencl /** This function performs a single time step in a Long Short-Term Memory (LSTM) layer. * @@ -120,13 +119,26 @@ public: * @param[in] projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within [-proj_clip, proj_clip]. * If set to 0.0f then clipping is disabled. */ - void configure(const ICLTensor *input, - const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - const ICLTensor *output_state_in, ICLTensor *cell_state_in, - ICLTensor *scratch_buffer, ICLTensor *output_state_out, ICLTensor *cell_state_out, ICLTensor *output, - const LSTMParams<ICLTensor> &lstm_params, const ActivationLayerInfo &activation_info, float cell_threshold = 0.f, float projection_threshold = 0.f); + void configure(const ICLTensor *input, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + const ICLTensor *output_state_in, + ICLTensor *cell_state_in, + ICLTensor *scratch_buffer, + ICLTensor *output_state_out, + ICLTensor *cell_state_out, + ICLTensor *output, + const LSTMParams<ICLTensor> &lstm_params, + const ActivationLayerInfo &activation_info, + float cell_threshold = 0.f, + float projection_threshold = 0.f); /** Initialize function's tensors. * * @param[in] compile_context The compile context to be used. @@ -166,13 +178,27 @@ public: * @param[in] projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within [-proj_clip, proj_clip]. * If set to 0.0f then clipping is disabled. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, - const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - const ICLTensor *output_state_in, ICLTensor *cell_state_in, - ICLTensor *scratch_buffer, ICLTensor *output_state_out, ICLTensor *cell_state_out, ICLTensor *output, - const LSTMParams<ICLTensor> &lstm_params, const ActivationLayerInfo &activation_info, float cell_threshold = 0.f, float projection_threshold = 0.f); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + const ICLTensor *output_state_in, + ICLTensor *cell_state_in, + ICLTensor *scratch_buffer, + ICLTensor *output_state_out, + ICLTensor *cell_state_out, + ICLTensor *output, + const LSTMParams<ICLTensor> &lstm_params, + const ActivationLayerInfo &activation_info, + float cell_threshold = 0.f, + float projection_threshold = 0.f); /** Static function to check if given info will lead to a valid configuration of @ref CLLSTMLayer * @@ -214,13 +240,26 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *output_state_in, const ITensorInfo *cell_state_in, - const ITensorInfo *scratch_buffer, const ITensorInfo *output_state_out, const ITensorInfo *cell_state_out, const ITensorInfo *output, - const LSTMParams<ITensorInfo> &lstm_params, const ActivationLayerInfo &activation_info, float cell_threshold = 0.f, float projection_threshold = 0.f); + static Status validate(const ITensorInfo *input, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_in, + const ITensorInfo *scratch_buffer, + const ITensorInfo *output_state_out, + const ITensorInfo *cell_state_out, + const ITensorInfo *output, + const LSTMParams<ITensorInfo> &lstm_params, + const ActivationLayerInfo &activation_info, + float cell_threshold = 0.f, + float projection_threshold = 0.f); // Inherited methods overridden: void run() override; @@ -311,7 +350,7 @@ private: bool _perform_projection_clipping; bool _is_prepared; bool _is_layer_norm_lstm; - const ICLTensor *_recurrent_to_cell_weights{ nullptr }; + const ICLTensor *_recurrent_to_cell_weights{nullptr}; }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLLSTMLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLLSTMLayerQuantized.h b/arm_compute/runtime/CL/functions/CLLSTMLayerQuantized.h index 9c004b85d0..8c116b1482 100644 --- a/arm_compute/runtime/CL/functions/CLLSTMLayerQuantized.h +++ b/arm_compute/runtime/CL/functions/CLLSTMLayerQuantized.h @@ -35,7 +35,6 @@ #include "arm_compute/runtime/CL/functions/CLQuantizationLayer.h" #include "arm_compute/runtime/CL/functions/CLSlice.h" #include "arm_compute/runtime/CL/functions/CLTranspose.h" - #include "arm_compute/runtime/common/LSTMParams.h" namespace arm_compute @@ -100,11 +99,22 @@ public: * @param[out] output_state_out Destination tensor. Output is a 2D tensor with dimensions [output_size, batch_size].Data types supported: Same as @p input. */ void configure(const ICLTensor *input, - const ICLTensor *input_to_input_weights, const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_input_weights, const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *input_gate_bias, const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - ICLTensor *cell_state_in, const ICLTensor *output_state_in, - ICLTensor *cell_state_out, ICLTensor *output_state_out); + const ICLTensor *input_to_input_weights, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_input_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *input_gate_bias, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + ICLTensor *cell_state_in, + const ICLTensor *output_state_in, + ICLTensor *cell_state_out, + ICLTensor *output_state_out); /** Initialize function's tensors. * * @param[in] compile_context The compile context to be used. @@ -126,12 +136,24 @@ public: * @param[out] cell_state_out Destination tensor. Output is a 2D tensor with dimensions [output_size, batch_size]. Data type supported: QSYMM16. * @param[out] output_state_out Destination tensor. Output is a 2D tensor with dimensions [output_size, batch_size].Data types supported: Same as @p input. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, - const ICLTensor *input_to_input_weights, const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_input_weights, const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *input_gate_bias, const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - ICLTensor *cell_state_in, const ICLTensor *output_state_in, - ICLTensor *cell_state_out, ICLTensor *output_state_out); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *input_to_input_weights, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_input_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *input_gate_bias, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + ICLTensor *cell_state_in, + const ICLTensor *output_state_in, + ICLTensor *cell_state_out, + ICLTensor *output_state_out); /** Static function to check if given info will lead to a valid configuration of @ref CLLSTMLayerQuantized * @@ -156,11 +178,22 @@ public: * @return a status */ static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_input_weights, const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_input_weights, const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *input_gate_bias, const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, - const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out); + const ITensorInfo *input_to_input_weights, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_input_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *input_gate_bias, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, + const ITensorInfo *output_state_out); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLLogicalAnd.h b/arm_compute/runtime/CL/functions/CLLogicalAnd.h index e3061e1dc3..4ff488782a 100644 --- a/arm_compute/runtime/CL/functions/CLLogicalAnd.h +++ b/arm_compute/runtime/CL/functions/CLLogicalAnd.h @@ -111,7 +111,8 @@ public: * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0. * @param[out] output Output tensor. Data types supported: same as @p input1. */ - void configure(const CLCompileContext &compile_context, ITensorInfo *input1, ITensorInfo *input2, ITensorInfo *output); + void + configure(const CLCompileContext &compile_context, ITensorInfo *input1, ITensorInfo *input2, ITensorInfo *output); /** Static function to check if given info will lead to a valid configuration of @ref arm_compute::opencl::kernels::ClLogicalBinaryKernel * * @param[in] input1 First tensor input info. Data types supported: U8. diff --git a/arm_compute/runtime/CL/functions/CLLogicalNot.h b/arm_compute/runtime/CL/functions/CLLogicalNot.h index 27fd0f9c9f..c7d9db93d7 100644 --- a/arm_compute/runtime/CL/functions/CLLogicalNot.h +++ b/arm_compute/runtime/CL/functions/CLLogicalNot.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLLOGICALNOT_H #define ARM_COMPUTE_CLLOGICALNOT_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -85,7 +84,7 @@ public: * @return a status */ static Status validate(const ITensorInfo *input, const ITensorInfo *output); - void run() override; + void run() override; private: struct Impl; diff --git a/arm_compute/runtime/CL/functions/CLLogicalOr.h b/arm_compute/runtime/CL/functions/CLLogicalOr.h index 893c22f721..64b6d83177 100644 --- a/arm_compute/runtime/CL/functions/CLLogicalOr.h +++ b/arm_compute/runtime/CL/functions/CLLogicalOr.h @@ -111,7 +111,8 @@ public: * The input tensor is [in, out] because its TensorInfo might be modified inside the kernel in case of broadcasting of dimension 0. * @param[out] output Output tensor. Data types supported: same as @p input1. */ - void configure(const CLCompileContext &compile_context, ITensorInfo *input1, ITensorInfo *input2, ITensorInfo *output); + void + configure(const CLCompileContext &compile_context, ITensorInfo *input1, ITensorInfo *input2, ITensorInfo *output); /** Static function to check if given info will lead to a valid configuration of @ref arm_compute::opencl::kernels::ClLogicalBinaryKernel * * @param[in] input1 First tensor input info. Data types supported: U8. diff --git a/arm_compute/runtime/CL/functions/CLMatMul.h b/arm_compute/runtime/CL/functions/CLMatMul.h index 9d54bab868..9c9939b9d0 100644 --- a/arm_compute/runtime/CL/functions/CLMatMul.h +++ b/arm_compute/runtime/CL/functions/CLMatMul.h @@ -27,6 +27,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -88,14 +89,23 @@ public: * @param[in] settings Contains flags for function level settings * @param[in] act_info (Optional) Contains activation function and lower and upper bound values for bounded activation functions. */ - void configure(const CLCompileContext &compile_context, ICLTensor *rhs, ICLTensor *lhs, ICLTensor *dst, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings = GpuMatMulSettings{}, const - ActivationLayerInfo &act_info = ActivationLayerInfo{}); + void configure(const CLCompileContext &compile_context, + ICLTensor *rhs, + ICLTensor *lhs, + ICLTensor *dst, + const MatMulInfo &matmul_info, + const GpuMatMulSettings &settings = GpuMatMulSettings{}, + const ActivationLayerInfo &act_info = ActivationLayerInfo{}); /** Initialise the kernel's inputs and output * * Similar to @ref CLMatMul::configure() */ - void configure(ICLTensor *lhs, ICLTensor *rhs, ICLTensor *dst, const MatMulInfo &matmul_info, const GpuMatMulSettings &settings = GpuMatMulSettings{}, const ActivationLayerInfo &act_info = - ActivationLayerInfo{}); + void configure(ICLTensor *lhs, + ICLTensor *rhs, + ICLTensor *dst, + const MatMulInfo &matmul_info, + const GpuMatMulSettings &settings = GpuMatMulSettings{}, + const ActivationLayerInfo &act_info = ActivationLayerInfo{}); /** Static function to check if given info will lead to a valid configuration of @ref CLMatMul. * * @@ -107,7 +117,11 @@ public: * @param[in] matmul_info Contains MatMul operation information described in @ref MatMulInfo. * @param[in] act_info (Optional) Contains activation function and lower and upper bound values for bounded activation functions. */ - static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *output, const MatMulInfo &matmul_info, const ActivationLayerInfo &act_info = ActivationLayerInfo{}); + static Status validate(const ITensorInfo *lhs, + const ITensorInfo *rhs, + const ITensorInfo *output, + const MatMulInfo &matmul_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo{}); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLMaxUnpoolingLayer.h b/arm_compute/runtime/CL/functions/CLMaxUnpoolingLayer.h index f7ff1234f6..2d2f064b4c 100644 --- a/arm_compute/runtime/CL/functions/CLMaxUnpoolingLayer.h +++ b/arm_compute/runtime/CL/functions/CLMaxUnpoolingLayer.h @@ -92,7 +92,11 @@ public: * @param[out] output Destination tensor. Data types supported: Same as @p input. * @param[in] pool_info Contains pooling operation information described in @ref PoolingLayerInfo. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *indices, ICLTensor *output, const PoolingLayerInfo &pool_info); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *indices, + ICLTensor *output, + const PoolingLayerInfo &pool_info); /** Static function to check if given info will lead to a valid configuration of @ref CLMaxUnpoolingLayer * * @param[in] input Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32. @@ -105,7 +109,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *indices, const ITensorInfo *output, const PoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *indices, + const ITensorInfo *output, + const PoolingLayerInfo &pool_info); // Inherited methods overridden: void run() override; @@ -114,5 +121,5 @@ private: CLFill _fill; std::unique_ptr<CLMaxUnpoolingLayerKernel> _unpooling_layer_kernel; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLMAXUNPOOLINGLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLMeanStdDevNormalizationLayer.h b/arm_compute/runtime/CL/functions/CLMeanStdDevNormalizationLayer.h index 68a7df24e6..951db3e419 100644 --- a/arm_compute/runtime/CL/functions/CLMeanStdDevNormalizationLayer.h +++ b/arm_compute/runtime/CL/functions/CLMeanStdDevNormalizationLayer.h @@ -65,7 +65,10 @@ public: * @param[out] output (Optional) Destination tensor. It can be nullptr in case of in-place computation. Data type supported: same as @p input * @param[in] epsilon (Optional) Small float to avoid division by zero in case of zero standard deviation. Defaults to 1e-8. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output = nullptr, float epsilon = 1e-8f); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output = nullptr, + float epsilon = 1e-8f); /** Static function to check if given info will lead to a valid configuration of @ref CLMeanStdDevNormalizationKernel * * @param[in] input Source tensor info with 2 dimensions. In case of @p output tensor info = nullptr, diff --git a/arm_compute/runtime/CL/functions/CLNormalizationLayer.h b/arm_compute/runtime/CL/functions/CLNormalizationLayer.h index 15406f7728..10fd8ed4c6 100644 --- a/arm_compute/runtime/CL/functions/CLNormalizationLayer.h +++ b/arm_compute/runtime/CL/functions/CLNormalizationLayer.h @@ -89,7 +89,10 @@ public: * Data types supported: same as @p input. Data layouts supported: same as @p input. * @param[in] norm_info Normalization layer information like the normalization type, normalization size and other parameters. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const NormalizationLayerInfo &norm_info); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + const NormalizationLayerInfo &norm_info); /** Static function to check if given info will lead to a valid configuration of @ref CLNormalizationLayer * * @param[in] input Source tensor. 3 lower dims represent a single input with dimensions [width, height, IFM], @@ -100,7 +103,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h b/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h index de5155c65a..3473af1004 100644 --- a/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h +++ b/arm_compute/runtime/CL/functions/CLNormalizePlanarYUVLayer.h @@ -62,7 +62,11 @@ public: * @param[in] std Standard deviation values tensor. 1 dimension with size equal to the number of input channels. * Data types supported: Same as @p input */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *mean, const ICLTensor *std); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const ICLTensor *mean, + const ICLTensor *std); /** Static function to check if given info will lead to a valid configuration of @ref CLNormalizePlanarYUVLayer * * @param[in] input Source tensor info. 3 lower dimensions represent a single input with dimensions [width, height, channels]. @@ -74,7 +78,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *mean, const ITensorInfo *std); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *mean, const ITensorInfo *std); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLNORMALIZEPLANARYUVLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLPadLayer.h b/arm_compute/runtime/CL/functions/CLPadLayer.h index 7f950bcfb3..89e693bd92 100644 --- a/arm_compute/runtime/CL/functions/CLPadLayer.h +++ b/arm_compute/runtime/CL/functions/CLPadLayer.h @@ -76,7 +76,11 @@ public: * @param[in] mode (Optional) Controls whether the padding should be filled with @p constant_value using CONSTANT, * or reflect the input, either including the border values (SYMMETRIC) or not (REFLECT). */ - void configure(ICLTensor *input, ICLTensor *output, const PaddingList &padding, PixelValue constant_value = PixelValue(), PaddingMode mode = PaddingMode::CONSTANT); + void configure(ICLTensor *input, + ICLTensor *output, + const PaddingList &padding, + PixelValue constant_value = PixelValue(), + PaddingMode mode = PaddingMode::CONSTANT); /** Initialize the function * * @param[in] compile_context The compile context to be used. @@ -88,8 +92,12 @@ public: * @param[in] mode (Optional) Controls whether the padding should be filled with @p constant_value using CONSTANT, * or reflect the input, either including the border values (SYMMETRIC) or not (REFLECT). */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const PaddingList &padding, PixelValue constant_value = PixelValue(), - PaddingMode mode = PaddingMode::CONSTANT); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + const PaddingList &padding, + PixelValue constant_value = PixelValue(), + PaddingMode mode = PaddingMode::CONSTANT); /** Static function to check if given info will lead to a valid configuration of @ref CLPadLayer. * @@ -101,7 +109,11 @@ public: * @param[in] mode (Optional) Controls whether the padding should be filled with @p constant_value using CONSTANT, * or reflect the input, either including the border values (SYMMETRIC) or not (REFLECT). */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PaddingList &padding, PixelValue constant_value = PixelValue(), PaddingMode mode = PaddingMode::CONSTANT); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const PaddingList &padding, + PixelValue constant_value = PixelValue(), + PaddingMode mode = PaddingMode::CONSTANT); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLPermute.h b/arm_compute/runtime/CL/functions/CLPermute.h index 8e15da2287..7ac0bf6b9c 100644 --- a/arm_compute/runtime/CL/functions/CLPermute.h +++ b/arm_compute/runtime/CL/functions/CLPermute.h @@ -78,7 +78,10 @@ public: * @param[in] output The output tensor. Data types supported: Same as @p input * @param[in] perm Permutation vector */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const PermutationVector &perm); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const PermutationVector &perm); /** Static function to check if given info will lead to a valid configuration of @ref CLPermute. * * @note Arbitrary permutation vectors are supported with rank not greater than 4 diff --git a/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h b/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h index 62b6d96ad5..f3e5cf9bd3 100644 --- a/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h +++ b/arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h @@ -84,8 +84,13 @@ public: * @param[in] rounding_policy Rounding policy. Supported rounding modes: to zero, to nearest even. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, float scale, - ConvertPolicy overflow_policy, RoundingPolicy rounding_policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + float scale, + ConvertPolicy overflow_policy, + RoundingPolicy rounding_policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output and convertion policy. * * @param[in] compile_context The compile context to be used. @@ -100,8 +105,14 @@ public: * @param[in] rounding_policy Rounding policy. Supported rounding modes: to zero, to nearest even. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, float scale, - ConvertPolicy overflow_policy, RoundingPolicy rounding_policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + float scale, + ConvertPolicy overflow_policy, + RoundingPolicy rounding_policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLPixelWiseMultiplication * * @param[in] input1 An input tensor info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/F32. @@ -115,8 +126,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, float scale, - ConvertPolicy overflow_policy, RoundingPolicy rounding_policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + float scale, + ConvertPolicy overflow_policy, + RoundingPolicy rounding_policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -151,7 +167,10 @@ public: * @param[out] output The output tensor, Data types supported: same as @p input1. Number of channels supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Initialise the kernel's inputs, output. * * @param[in] compile_context The compile context to be used. @@ -162,7 +181,11 @@ public: * @param[out] output The output tensor, Data types supported: same as @p input1. Number of channels supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input1, ICLTensor *input2, ICLTensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const CLCompileContext &compile_context, + ICLTensor *input1, + ICLTensor *input2, + ICLTensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CLComplexPixelWiseMultiplication * * @param[in] input1 An input tensor info. Data types supported: F16/F32. Number of channels supported: 2. @@ -170,7 +193,10 @@ public: * @param[in] output The output tensor info, Data types supported: same as @p input1. Number of channels supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLPooling3dLayer.h b/arm_compute/runtime/CL/functions/CLPooling3dLayer.h index 91c46770da..1c69148771 100644 --- a/arm_compute/runtime/CL/functions/CLPooling3dLayer.h +++ b/arm_compute/runtime/CL/functions/CLPooling3dLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLPOOLING3DLAYER_H #define ARM_COMPUTE_CLPOOLING3DLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -82,7 +81,10 @@ public: * @param[out] output Destination tensor. Data types supported: Same as @p input. * @param[in] pool_info Contains 3d pooling operation information described in @ref Pooling3dLayerInfo. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const Pooling3dLayerInfo &pool_info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const Pooling3dLayerInfo &pool_info); /** Static function to check if given info will lead to a valid configuration of @ref CLPooling3dLayer * * @param[in] input Source tensor info. Data types supported: F16/F32/QASYMM8/QASYMM8_SIGNED. diff --git a/arm_compute/runtime/CL/functions/CLPoolingLayer.h b/arm_compute/runtime/CL/functions/CLPoolingLayer.h index 2163c16801..3dbdf8aeea 100644 --- a/arm_compute/runtime/CL/functions/CLPoolingLayer.h +++ b/arm_compute/runtime/CL/functions/CLPoolingLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CLPOOLINGLAYER_H #define ARM_COMPUTE_CLPOOLINGLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -74,7 +73,8 @@ public: * @param[in] pool_info Contains pooling operation information described in @ref PoolingLayerInfo. * @param[out] indices (optional) The indices of the maximal values. Data type supported: U32. */ - void configure(ICLTensor *input, ICLTensor *output, const PoolingLayerInfo &pool_info, ICLTensor *indices = nullptr); + void + configure(ICLTensor *input, ICLTensor *output, const PoolingLayerInfo &pool_info, ICLTensor *indices = nullptr); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -83,7 +83,11 @@ public: * @param[in] pool_info Contains pooling operation information described in @ref PoolingLayerInfo. * @param[out] indices (optional) The indices of the maximal values. Data type supported: U32. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const PoolingLayerInfo &pool_info, ICLTensor *indices = nullptr); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + const PoolingLayerInfo &pool_info, + ICLTensor *indices = nullptr); /** Static function to check if given info will lead to a valid configuration of @ref CLPoolingLayer * * @param[in] input Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32. @@ -93,7 +97,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PoolingLayerInfo &pool_info, const ITensorInfo *indices = nullptr); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const PoolingLayerInfo &pool_info, + const ITensorInfo *indices = nullptr); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLPriorBoxLayer.h b/arm_compute/runtime/CL/functions/CLPriorBoxLayer.h index 9b36c9e433..4ede906baa 100644 --- a/arm_compute/runtime/CL/functions/CLPriorBoxLayer.h +++ b/arm_compute/runtime/CL/functions/CLPriorBoxLayer.h @@ -66,7 +66,11 @@ public: * @param[out] output Destination tensor. Output dimensions are [W * H * num_priors * 4, 2]. Data types and layouts supported: same as @p input1 * @param[in] info Prior box layer info. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input1, const ICLTensor *input2, ICLTensor *output, const PriorBoxLayerInfo &info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input1, + const ICLTensor *input2, + ICLTensor *output, + const PriorBoxLayerInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLPriorBoxLayer * * @param[in] input1 First source tensor info. Data types supported: F32. Data layouts supported: NCHW/NHWC. @@ -76,12 +80,15 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const PriorBoxLayerInfo &info); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const PriorBoxLayerInfo &info); private: cl::Buffer _min; cl::Buffer _max; cl::Buffer _aspect_ratios; }; -} // arm_compute +} // namespace arm_compute #endif /* ARM_COMPUTE_CLPRIORBOXLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLQLSTMLayer.h b/arm_compute/runtime/CL/functions/CLQLSTMLayer.h index 1b0b759d74..3e76da086f 100644 --- a/arm_compute/runtime/CL/functions/CLQLSTMLayer.h +++ b/arm_compute/runtime/CL/functions/CLQLSTMLayer.h @@ -32,7 +32,6 @@ #include "arm_compute/runtime/CL/functions/CLGEMMLowpOutputStage.h" #include "arm_compute/runtime/CL/functions/CLPixelWiseMultiplication.h" #include "arm_compute/runtime/CL/functions/CLTranspose.h" - #include "arm_compute/runtime/common/LSTMParams.h" namespace arm_compute @@ -127,12 +126,21 @@ public: * projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. */ - void configure(const ICLTensor *input, - const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - ICLTensor *cell_state_in, ICLTensor *output_state_in, - ICLTensor *cell_state_out, ICLTensor *output_state_out, ICLTensor *output, + void configure(const ICLTensor *input, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + ICLTensor *cell_state_in, + ICLTensor *output_state_in, + ICLTensor *cell_state_out, + ICLTensor *output_state_out, + ICLTensor *output, const LSTMParams<ICLTensor> &lstm_params); /** Initialize function's tensors. @@ -177,12 +185,22 @@ public: * projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, - const ICLTensor *input_to_forget_weights, const ICLTensor *input_to_cell_weights, const ICLTensor *input_to_output_weights, - const ICLTensor *recurrent_to_forget_weights, const ICLTensor *recurrent_to_cell_weights, const ICLTensor *recurrent_to_output_weights, - const ICLTensor *forget_gate_bias, const ICLTensor *cell_bias, const ICLTensor *output_gate_bias, - ICLTensor *cell_state_in, ICLTensor *output_state_in, - ICLTensor *cell_state_out, ICLTensor *output_state_out, ICLTensor *output, + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *input_to_forget_weights, + const ICLTensor *input_to_cell_weights, + const ICLTensor *input_to_output_weights, + const ICLTensor *recurrent_to_forget_weights, + const ICLTensor *recurrent_to_cell_weights, + const ICLTensor *recurrent_to_output_weights, + const ICLTensor *forget_gate_bias, + const ICLTensor *cell_bias, + const ICLTensor *output_gate_bias, + ICLTensor *cell_state_in, + ICLTensor *output_state_in, + ICLTensor *cell_state_out, + ICLTensor *output_state_out, + ICLTensor *output, const LSTMParams<ICLTensor> &lstm_params); /** Static function to check if given info will lead to a valid configuration of @ref CLQLSTMLayer @@ -227,12 +245,21 @@ public: * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. * @return a status */ - static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, - const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out, const ITensorInfo *output, + static Status validate(const ITensorInfo *input, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, + const ITensorInfo *output_state_out, + const ITensorInfo *output, const LSTMParams<ITensorInfo> &lstm_params); // Inherited methods overridden: @@ -266,10 +293,18 @@ private: * @param[in] mm_res_info Tensor info to be used to initialize output stage result tensor. * */ - void configure_mm(const CLCompileContext &compile_context, CLGEMMLowpMatrixMultiplyCore &mm, CLGEMMLowpOutputStage &outstage, GEMMLowpOutputStageInfo &gemmlowp_info, - const ICLTensor *mm_input, const ICLTensor *mm_weights, const ICLTensor *bias, CLTensor *mm_res, - CLTensor *outstage_res, float gemmlowp_scale, - const TensorInfo &mm_res_info, const TensorInfo &outstage_tensor_info); + void configure_mm(const CLCompileContext &compile_context, + CLGEMMLowpMatrixMultiplyCore &mm, + CLGEMMLowpOutputStage &outstage, + GEMMLowpOutputStageInfo &gemmlowp_info, + const ICLTensor *mm_input, + const ICLTensor *mm_weights, + const ICLTensor *bias, + CLTensor *mm_res, + CLTensor *outstage_res, + float gemmlowp_scale, + const TensorInfo &mm_res_info, + const TensorInfo &outstage_tensor_info); MemoryGroup _memory_group{}; @@ -278,8 +313,8 @@ private: { static constexpr uint32_t max_dimension_supported = 2; - ICLTensor *_src{ nullptr }; - ICLTensor *_dst{ nullptr }; + ICLTensor *_src{nullptr}; + ICLTensor *_dst{nullptr}; size_t _row_size{}; Window _window{}; @@ -368,7 +403,7 @@ private: CLArithmeticAddition _accumulate_projection{}; CLActivationLayer _projection_clip{}; std::array<std::unique_ptr<CLQLSTMLayerNormalizationKernel>, _layer_norm_count> _layer_norms; - CLCopy _copy_output; + CLCopy _copy_output; TensorCopyKernel _projection_bias_copy{}; TensorCopyKernel _projection_output_to_accumulate_copy{}; @@ -376,21 +411,18 @@ private: TensorCopyKernel _hidden_to_output_copy{}; // Tensor pointers - const ICLTensor *_input_to_input_weights - { - nullptr - }; - const ICLTensor *_recurrent_to_input_weights{ nullptr }; - const ICLTensor *_projection_bias{ nullptr }; - const ICLTensor *_input_to_forget_weights{ nullptr }; - const ICLTensor *_input_to_cell_weights{ nullptr }; - const ICLTensor *_input_to_output_weights{ nullptr }; - const ICLTensor *_recurrent_to_forget_weights{ nullptr }; - const ICLTensor *_recurrent_to_cell_weights{ nullptr }; - const ICLTensor *_recurrent_to_output_weights{ nullptr }; - const ICLTensor *_projection_weights{ nullptr }; - std::array<const ICLTensor *, _layer_norm_count> _layer_norm_weights{ {} }; - std::array<const ICLTensor *, _layer_norm_count> _layer_norm_bias{ {} }; + const ICLTensor *_input_to_input_weights{nullptr}; + const ICLTensor *_recurrent_to_input_weights{nullptr}; + const ICLTensor *_projection_bias{nullptr}; + const ICLTensor *_input_to_forget_weights{nullptr}; + const ICLTensor *_input_to_cell_weights{nullptr}; + const ICLTensor *_input_to_output_weights{nullptr}; + const ICLTensor *_recurrent_to_forget_weights{nullptr}; + const ICLTensor *_recurrent_to_cell_weights{nullptr}; + const ICLTensor *_recurrent_to_output_weights{nullptr}; + const ICLTensor *_projection_weights{nullptr}; + std::array<const ICLTensor *, _layer_norm_count> _layer_norm_weights{{}}; + std::array<const ICLTensor *, _layer_norm_count> _layer_norm_bias{{}}; using LayerNormIndexType = typename std::underlying_type<LayerNormGate>::type; inline LayerNormIndexType getGateIndex(LayerNormGate g) @@ -423,78 +455,78 @@ private: return *_layer_norms[getGateIndex(g)]; } - inline void configure_layer_norm(LayerNormGate g, const ICLTensor *in); + inline void configure_layer_norm(LayerNormGate g, const ICLTensor *in); inline static Status validate_layer_norm(const ITensorInfo &in, const ITensorInfo &weight, const ITensorInfo &bias); // Temporary tensors - CLTensor _input_to_forget_weights_transposed{ nullptr }; - CLTensor _input_to_cell_weights_transposed{ nullptr }; - CLTensor _input_to_output_weights_transposed{ nullptr }; - CLTensor _input_to_input_weights_transposed{ nullptr }; - CLTensor _recurrent_to_forget_weights_transposed{ nullptr }; - CLTensor _recurrent_to_cell_weights_transposed{ nullptr }; - CLTensor _recurrent_to_output_weights_transposed{ nullptr }; - CLTensor _recurrent_to_input_weights_transposed{ nullptr }; - CLTensor _projection_weights_transposed{ nullptr }; - CLTensor _input_to_input_eff_bias{ nullptr }; - CLTensor _recurrent_to_input_eff_bias{ nullptr }; - CLTensor _input_to_forget_eff_bias{ nullptr }; - CLTensor _recurrent_to_forget_eff_bias{ nullptr }; - CLTensor _input_to_cell_eff_bias{ nullptr }; - CLTensor _recurrent_to_cell_eff_bias{ nullptr }; - CLTensor _input_to_output_eff_bias{ nullptr }; - CLTensor _recurrent_to_output_eff_bias{ nullptr }; - CLTensor _projection_reduction_res{ nullptr }; - CLTensor _projection_eff_bias{ nullptr }; - CLTensor _mm_input_to_forget_res{ nullptr }; - CLTensor _mm_recurrent_to_forget_res{ nullptr }; - CLTensor _mul_cell_to_forget_res{ nullptr }; - CLTensor _input_to_forget_outstage_res{ nullptr }; - CLTensor _cell_to_forget_outstage_res{ nullptr }; - CLTensor _recurrent_to_forget_outstage_res{ nullptr }; - CLTensor _forget_gate{ nullptr }; - CLTensor _mm_input_to_cell_res{ nullptr }; - CLTensor _input_to_cell_outstage_res{ nullptr }; - CLTensor _mm_recurrent_to_cell_res{ nullptr }; - CLTensor _recurrent_to_cell_outstage_res{ nullptr }; - CLTensor _cell_gate{ nullptr }; - CLTensor _mul_input_cell_res{ nullptr }; - CLTensor _mm_input_to_input_res{ nullptr }; - CLTensor _input_to_input_outstage_res{ nullptr }; - CLTensor _mm_recurrent_to_input_res{ nullptr }; - CLTensor _mul_cell_to_input_res{ nullptr }; - CLTensor _cell_to_input_outstage_res{ nullptr }; - CLTensor _recurrent_to_input_outstage_res{ nullptr }; - CLTensor _input_gate{ nullptr }; - CLTensor _mm_input_to_output_res{ nullptr }; - CLTensor _input_to_output_outstage_res{ nullptr }; - CLTensor _mm_recurrent_to_output_res{ nullptr }; - CLTensor _mul_cell_to_output_res{ nullptr }; - CLTensor _cell_to_output_outstage_res{ nullptr }; - CLTensor _recurrent_to_output_outstage_res{ nullptr }; - CLTensor _output_gate{ nullptr }; - CLTensor _hidden_mul_res{ nullptr }; - CLTensor _hidden_gate{ nullptr }; - CLTensor _mm_projection_res{ nullptr }; - CLTensor _projection_outstage_res{ nullptr }; - CLTensor _projection_out_res{ nullptr }; - CLTensor _projection_accumulate_res{ nullptr }; - CLTensor _ones{ nullptr }; - std::array<CLTensor, _layer_norm_count> _layer_norm_output{ {} }; + CLTensor _input_to_forget_weights_transposed{nullptr}; + CLTensor _input_to_cell_weights_transposed{nullptr}; + CLTensor _input_to_output_weights_transposed{nullptr}; + CLTensor _input_to_input_weights_transposed{nullptr}; + CLTensor _recurrent_to_forget_weights_transposed{nullptr}; + CLTensor _recurrent_to_cell_weights_transposed{nullptr}; + CLTensor _recurrent_to_output_weights_transposed{nullptr}; + CLTensor _recurrent_to_input_weights_transposed{nullptr}; + CLTensor _projection_weights_transposed{nullptr}; + CLTensor _input_to_input_eff_bias{nullptr}; + CLTensor _recurrent_to_input_eff_bias{nullptr}; + CLTensor _input_to_forget_eff_bias{nullptr}; + CLTensor _recurrent_to_forget_eff_bias{nullptr}; + CLTensor _input_to_cell_eff_bias{nullptr}; + CLTensor _recurrent_to_cell_eff_bias{nullptr}; + CLTensor _input_to_output_eff_bias{nullptr}; + CLTensor _recurrent_to_output_eff_bias{nullptr}; + CLTensor _projection_reduction_res{nullptr}; + CLTensor _projection_eff_bias{nullptr}; + CLTensor _mm_input_to_forget_res{nullptr}; + CLTensor _mm_recurrent_to_forget_res{nullptr}; + CLTensor _mul_cell_to_forget_res{nullptr}; + CLTensor _input_to_forget_outstage_res{nullptr}; + CLTensor _cell_to_forget_outstage_res{nullptr}; + CLTensor _recurrent_to_forget_outstage_res{nullptr}; + CLTensor _forget_gate{nullptr}; + CLTensor _mm_input_to_cell_res{nullptr}; + CLTensor _input_to_cell_outstage_res{nullptr}; + CLTensor _mm_recurrent_to_cell_res{nullptr}; + CLTensor _recurrent_to_cell_outstage_res{nullptr}; + CLTensor _cell_gate{nullptr}; + CLTensor _mul_input_cell_res{nullptr}; + CLTensor _mm_input_to_input_res{nullptr}; + CLTensor _input_to_input_outstage_res{nullptr}; + CLTensor _mm_recurrent_to_input_res{nullptr}; + CLTensor _mul_cell_to_input_res{nullptr}; + CLTensor _cell_to_input_outstage_res{nullptr}; + CLTensor _recurrent_to_input_outstage_res{nullptr}; + CLTensor _input_gate{nullptr}; + CLTensor _mm_input_to_output_res{nullptr}; + CLTensor _input_to_output_outstage_res{nullptr}; + CLTensor _mm_recurrent_to_output_res{nullptr}; + CLTensor _mul_cell_to_output_res{nullptr}; + CLTensor _cell_to_output_outstage_res{nullptr}; + CLTensor _recurrent_to_output_outstage_res{nullptr}; + CLTensor _output_gate{nullptr}; + CLTensor _hidden_mul_res{nullptr}; + CLTensor _hidden_gate{nullptr}; + CLTensor _mm_projection_res{nullptr}; + CLTensor _projection_outstage_res{nullptr}; + CLTensor _projection_out_res{nullptr}; + CLTensor _projection_accumulate_res{nullptr}; + CLTensor _ones{nullptr}; + std::array<CLTensor, _layer_norm_count> _layer_norm_output{{}}; inline CLTensor &get_layer_norm_output(LayerNormGate g) { return _layer_norm_output[getGateIndex(g)]; } - bool _is_prepared{ false }; - bool _has_cifg{ false }; - bool _has_cell_clipping{ false }; - bool _has_projection{ false }; - bool _has_projection_clipping{ false }; - bool _has_peephole{ false }; - bool _has_layer_norm{ false }; - bool _projection_tensor_copy_required{ false }; + bool _is_prepared{false}; + bool _has_cifg{false}; + bool _has_cell_clipping{false}; + bool _has_projection{false}; + bool _has_projection_clipping{false}; + bool _has_peephole{false}; + bool _has_layer_norm{false}; + bool _projection_tensor_copy_required{false}; }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLQLSTMLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLRNNLayer.h b/arm_compute/runtime/CL/functions/CLRNNLayer.h index 2b3b35e37d..a8d835d04d 100644 --- a/arm_compute/runtime/CL/functions/CLRNNLayer.h +++ b/arm_compute/runtime/CL/functions/CLRNNLayer.h @@ -24,12 +24,12 @@ #ifndef ARM_COMPUTE_CLRNN_LAYER_H #define ARM_COMPUTE_CLRNN_LAYER_H -#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include "arm_compute/runtime/CL/functions/CLActivationLayer.h" #include "arm_compute/runtime/CL/functions/CLCopy.h" #include "arm_compute/runtime/CL/functions/CLElementwiseOperations.h" #include "arm_compute/runtime/CL/functions/CLFullyConnectedLayer.h" #include "arm_compute/runtime/CL/functions/CLGEMM.h" +#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include <memory> @@ -69,7 +69,13 @@ public: * @param[in,out] hidden_state Output tensor of shape [num_units, batch_size]. Data types supported: Same as @p input * @param[in] info Activation layer parameter. */ - void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *recurrent_weights, const ICLTensor *bias, ICLTensor *hidden_state, ICLTensor *output, ActivationLayerInfo &info); + void configure(const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *recurrent_weights, + const ICLTensor *bias, + ICLTensor *hidden_state, + ICLTensor *output, + ActivationLayerInfo &info); /** Initialize the function * * @param[in] compile_context The compile context to be used. @@ -81,8 +87,14 @@ public: * @param[in,out] hidden_state Output tensor of shape [num_units, batch_size]. Data types supported: Same as @p input * @param[in] info Activation layer parameter. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *recurrent_weights, const ICLTensor *bias, ICLTensor *hidden_state, - ICLTensor *output, ActivationLayerInfo &info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *recurrent_weights, + const ICLTensor *bias, + ICLTensor *hidden_state, + ICLTensor *output, + ActivationLayerInfo &info); /** Initialize the function * * @param[in] input Input is a 2-D tensor of shape [input_size, batch_size]. Data types supported: F16/F32 @@ -95,7 +107,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *recurrent_weights, const ITensorInfo *bias, const ITensorInfo *hidden_state, const ITensorInfo *output, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *recurrent_weights, + const ITensorInfo *bias, + const ITensorInfo *hidden_state, + const ITensorInfo *output, const ActivationLayerInfo &info); // Inherited methods overridden: @@ -114,5 +131,5 @@ private: CLTensor _add_output; bool _is_prepared; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLRNN_LAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLROIAlignLayer.h b/arm_compute/runtime/CL/functions/CLROIAlignLayer.h index 1eaea1b297..14d3476711 100644 --- a/arm_compute/runtime/CL/functions/CLROIAlignLayer.h +++ b/arm_compute/runtime/CL/functions/CLROIAlignLayer.h @@ -68,7 +68,8 @@ public: * @note The z dimensions of @p output tensor and @p input tensor must be the same. * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. */ - void configure(const ICLTensor *input, const ICLTensor *rois, ICLTensor *output, const ROIPoolingLayerInfo &pool_info); + void + configure(const ICLTensor *input, const ICLTensor *rois, ICLTensor *output, const ROIPoolingLayerInfo &pool_info); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -84,7 +85,11 @@ public: * @note The z dimensions of @p output tensor and @p input tensor must be the same. * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *rois, ICLTensor *output, const ROIPoolingLayerInfo &pool_info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *rois, + ICLTensor *output, + const ROIPoolingLayerInfo &pool_info); /** Static function to check if given info will lead to a valid configuration of @ref CLROIAlignLayer * * @param[in] input Source tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32. @@ -100,7 +105,10 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *rois, ITensorInfo *output, const ROIPoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *rois, + ITensorInfo *output, + const ROIPoolingLayerInfo &pool_info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLROIALIGNLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLROIPoolingLayer.h b/arm_compute/runtime/CL/functions/CLROIPoolingLayer.h index 151586a1f6..86294596d2 100644 --- a/arm_compute/runtime/CL/functions/CLROIPoolingLayer.h +++ b/arm_compute/runtime/CL/functions/CLROIPoolingLayer.h @@ -66,7 +66,8 @@ public: * @note The z dimensions of @p output tensor and @p input tensor must be the same. * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. */ - void configure(const ICLTensor *input, const ICLTensor *rois, ICLTensor *output, const ROIPoolingLayerInfo &pool_info); + void + configure(const ICLTensor *input, const ICLTensor *rois, ICLTensor *output, const ROIPoolingLayerInfo &pool_info); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -81,7 +82,11 @@ public: * @note The z dimensions of @p output tensor and @p input tensor must be the same. * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *rois, const ICLTensor *output, const ROIPoolingLayerInfo &pool_info); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *rois, + const ICLTensor *output, + const ROIPoolingLayerInfo &pool_info); /** Static function to check if given info will lead to a valid configuration of @ref CLROIPoolingLayer * @@ -97,7 +102,10 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *rois, ITensorInfo *output, const ROIPoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *rois, + ITensorInfo *output, + const ROIPoolingLayerInfo &pool_info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CLROIPOOLINGLAYER_H */ diff --git a/arm_compute/runtime/CL/functions/CLRange.h b/arm_compute/runtime/CL/functions/CLRange.h index fbce05162c..ed665bc398 100644 --- a/arm_compute/runtime/CL/functions/CLRange.h +++ b/arm_compute/runtime/CL/functions/CLRange.h @@ -73,7 +73,8 @@ public: * @param[in] end The ending (not including) value of the sequence. * @param[in] step The gap between each pair of values in the sequence. Default is 1. */ - void configure(const CLCompileContext &compile_context, ICLTensor *output, float start, float end, float step = 1.f); + void + configure(const CLCompileContext &compile_context, ICLTensor *output, float start, float end, float step = 1.f); /** Static function to check if given info will lead to a valid configuration of @ref CLRange * * @param[in] output Output tensor info. Data types supported: U8/S8/QASYMM8/U16/S16/U32/S32/F16/F32. diff --git a/arm_compute/runtime/CL/functions/CLReduceMean.h b/arm_compute/runtime/CL/functions/CLReduceMean.h index 1ce088b2ce..640fe7cf1b 100644 --- a/arm_compute/runtime/CL/functions/CLReduceMean.h +++ b/arm_compute/runtime/CL/functions/CLReduceMean.h @@ -24,12 +24,12 @@ #ifndef ARM_COMPUTE_CL_REDUCE_MEAN_H #define ARM_COMPUTE_CL_REDUCE_MEAN_H -#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include "arm_compute/runtime/CL/functions/CLDequantizationLayer.h" #include "arm_compute/runtime/CL/functions/CLElementwiseOperations.h" #include "arm_compute/runtime/CL/functions/CLQuantizationLayer.h" #include "arm_compute/runtime/CL/functions/CLReductionOperation.h" #include "arm_compute/runtime/CL/functions/CLReshapeLayer.h" +#include "arm_compute/runtime/CL/ICLSimpleFunction.h" #include "arm_compute/runtime/IMemoryManager.h" namespace arm_compute @@ -74,7 +74,11 @@ public: * @param[in] keep_dims If positive, retains reduced dimensions with length 1. * @param[out] output Destination tensor. Data type supported: Same as @p input */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const Coordinates &reduction_axis, bool keep_dims, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const Coordinates &reduction_axis, + bool keep_dims, + ICLTensor *output); /** Static function to check if given info will lead to a valid configuration of @ref CLReduceMean * @@ -85,7 +89,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const Coordinates &reduction_axis, bool keep_dims, const ITensorInfo *output); + static Status + validate(const ITensorInfo *input, const Coordinates &reduction_axis, bool keep_dims, const ITensorInfo *output); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLReductionOperation.h b/arm_compute/runtime/CL/functions/CLReductionOperation.h index 2245735b62..80068ac35c 100644 --- a/arm_compute/runtime/CL/functions/CLReductionOperation.h +++ b/arm_compute/runtime/CL/functions/CLReductionOperation.h @@ -80,7 +80,8 @@ public: * @param[in] op Reduction operation to perform. Operations supported: MEAN_SUM, PROD, SUM_SQUARE, SUM, MIN, MAX * @param[in] keep_dims (Optional) Whether to keep the reduced dimension after the operation. Defaults to true. */ - void configure(ICLTensor *input, ICLTensor *output, unsigned int axis, ReductionOperation op, bool keep_dims = true); + void + configure(ICLTensor *input, ICLTensor *output, unsigned int axis, ReductionOperation op, bool keep_dims = true); /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. @@ -90,7 +91,12 @@ public: * @param[in] op Reduction operation to perform. Operations supported: MEAN_SUM, PROD, SUM_SQUARE, SUM, MIN, MAX * @param[in] keep_dims (Optional) Whether to keep the reduced dimension after the operation. Defaults to true. */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, unsigned int axis, ReductionOperation op, bool keep_dims = true); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + unsigned int axis, + ReductionOperation op, + bool keep_dims = true); /** Static function to check if given info will lead to a valid configuration of @ref CLReductionOperation. * @@ -102,7 +108,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, unsigned int axis, ReductionOperation op, bool keep_dims = true); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + unsigned int axis, + ReductionOperation op, + bool keep_dims = true); // Inherited methods overridden: void run() override; @@ -118,4 +128,4 @@ private: bool _is_reshape_required; }; } // namespace arm_compute -#endif /* ARM_COMPUTE_CLREDUCTIONOPERATION_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_CLREDUCTIONOPERATION_H */ diff --git a/arm_compute/runtime/CL/functions/CLReshapeLayer.h b/arm_compute/runtime/CL/functions/CLReshapeLayer.h index 7346b65e9b..dad90e6ba9 100644 --- a/arm_compute/runtime/CL/functions/CLReshapeLayer.h +++ b/arm_compute/runtime/CL/functions/CLReshapeLayer.h @@ -26,6 +26,7 @@ #include "arm_compute/runtime/CL/ICLOperator.h" #include "arm_compute/runtime/CL/ICLSimpleFunction.h" + #include <memory> namespace arm_compute diff --git a/arm_compute/runtime/CL/functions/CLReverse.h b/arm_compute/runtime/CL/functions/CLReverse.h index 94c63ca92d..0defc3f28b 100644 --- a/arm_compute/runtime/CL/functions/CLReverse.h +++ b/arm_compute/runtime/CL/functions/CLReverse.h @@ -59,7 +59,10 @@ public: * @param[out] output Output tensor. Data type supported: Same as @p input * @param[in] axis Axis tensor. Contains the indices of the dimensions to reverse. Data type supported: U32 */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const ICLTensor *axis); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const ICLTensor *axis); /** Static function to check if given info will lead to a valid configuration of @ref CLReverseKernel * * @param[in] input Input tensor info. Data types supported: All. diff --git a/arm_compute/runtime/CL/functions/CLScale.h b/arm_compute/runtime/CL/functions/CLScale.h index ddb4a23531..5c3824eb58 100644 --- a/arm_compute/runtime/CL/functions/CLScale.h +++ b/arm_compute/runtime/CL/functions/CLScale.h @@ -83,7 +83,10 @@ public: * All but the lowest two dimensions must be the same size as in the input tensor, i.e. scaling is only performed within the XY-plane. * @param[in] info @ref ScaleKernelInfo descriptor to be used to configure */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, ICLTensor *output, const ScaleKernelInfo &info); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + ICLTensor *output, + const ScaleKernelInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLScale * diff --git a/arm_compute/runtime/CL/functions/CLSelect.h b/arm_compute/runtime/CL/functions/CLSelect.h index 8b1e6b2019..effcb58313 100644 --- a/arm_compute/runtime/CL/functions/CLSelect.h +++ b/arm_compute/runtime/CL/functions/CLSelect.h @@ -62,7 +62,11 @@ public: * @param[in] y Second input tensor. Data types supported: Same as @p x * @param[out] output Output tensor. Data types supported: Same as @p x. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *c, const ICLTensor *x, const ICLTensor *y, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *c, + const ICLTensor *x, + const ICLTensor *y, + ICLTensor *output); /** Static function to check if given info will lead to a valid configuration of @ref CLSelect * * @param[in] c Condition input tensor. Data types supported: U8. diff --git a/arm_compute/runtime/CL/functions/CLSlice.h b/arm_compute/runtime/CL/functions/CLSlice.h index 297bcd86fe..7a274ded72 100644 --- a/arm_compute/runtime/CL/functions/CLSlice.h +++ b/arm_compute/runtime/CL/functions/CLSlice.h @@ -84,7 +84,11 @@ public: * @param[in] starts The starts of the dimensions of the input tensor to be sliced. The length must be of rank(input). * @param[in] ends The ends of the dimensions of the input tensor to be sliced. The length must be of rank(input). */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const Coordinates &starts, const Coordinates &ends); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const Coordinates &starts, + const Coordinates &ends); /** Static function to check if given info will lead to a valid configuration of @ref CLSlice * @@ -100,7 +104,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); // Inherited methods overridden: void run() override; @@ -129,7 +134,11 @@ public: * @param[in] starts The starts of the dimensions of the input tensor to be sliced. The length must be of rank(input). * @param[in] ends The ends of the dimensions of the input tensor to be sliced. The length must be of rank(input). */ - void configure(const CLCompileContext &compile_context, const ITensorInfo *input, ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); + void configure(const CLCompileContext &compile_context, + const ITensorInfo *input, + ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends); /** Static function to check if given info will lead to a valid configuration of @ref CLSlice * @@ -145,7 +154,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); }; } // namespace experimental } // namespace arm_compute diff --git a/arm_compute/runtime/CL/functions/CLSoftmaxLayer.h b/arm_compute/runtime/CL/functions/CLSoftmaxLayer.h index 687f8ff6d8..70ef1f4402 100644 --- a/arm_compute/runtime/CL/functions/CLSoftmaxLayer.h +++ b/arm_compute/runtime/CL/functions/CLSoftmaxLayer.h @@ -87,7 +87,11 @@ public: * @param[in] axis (Optional) The dimension in which to apply the function. E.g. for input of shape 4x5x6 and * axis=1, softmax will be applied to 4x6=24 vectors of size 5. Defaults to 0 */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, float beta = 1.0f, int32_t axis = 0); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + float beta = 1.0f, + int32_t axis = 0); /** Static function to check if given info will lead to a valid configuration of @ref CLSoftmaxLayer * * @param[in] input Source tensor. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32 for Softmax and F16/F32 for Log Softmax diff --git a/arm_compute/runtime/CL/functions/CLSpaceToBatchLayer.h b/arm_compute/runtime/CL/functions/CLSpaceToBatchLayer.h index 304a74137e..191f4863d5 100644 --- a/arm_compute/runtime/CL/functions/CLSpaceToBatchLayer.h +++ b/arm_compute/runtime/CL/functions/CLSpaceToBatchLayer.h @@ -83,7 +83,11 @@ public: * @param[in] paddings 2-D tensor with shape [2, M] (First dimension is the fastest-changing dimension). Supported M: 2. Data types supported: S32 * @param[out] output Tensor output. Data types supported: same as @p input */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *block_shape, const ICLTensor *paddings, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const ICLTensor *block_shape, + const ICLTensor *paddings, + ICLTensor *output); /** Set the input and output tensors. (Static block shape and paddings) * * @param[in] input Tensor input. Supported tensor rank: 4. Data types supported: All. @@ -93,7 +97,12 @@ public: * @param[in] padding_right The padding at the end of every dimension of the output tensor. * @param[out] output Tensor output. Data types supported: same as @p input */ - void configure(const ICLTensor *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, ICLTensor *output); + void configure(const ICLTensor *input, + const int block_shape_x, + const int block_shape_y, + const Size2D &padding_left, + const Size2D &padding_right, + ICLTensor *output); /** Set the input and output tensors. (Static block shape and paddings) * * @param[in] compile_context The compile context to be used. @@ -104,8 +113,13 @@ public: * @param[in] padding_right The padding at the end of every dimension of the output tensor. * @param[out] output Tensor output. Data types supported: same as @p input */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, - ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const int block_shape_x, + const int block_shape_y, + const Size2D &padding_left, + const Size2D &padding_right, + ICLTensor *output); /** Static function to check if given info will lead to a valid configuration of @ref CLSpaceToBatchLayer * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -115,7 +129,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *block_shape, const ITensorInfo *paddings, const ITensorInfo *output); + static Status validate(const ITensorInfo *input, + const ITensorInfo *block_shape, + const ITensorInfo *paddings, + const ITensorInfo *output); /** Static function to check if given info will lead to a valid configuration of @ref CLSpaceToBatchLayer (Static block shape and paddings) * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -127,7 +144,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, const ITensorInfo *output); + static Status validate(const ITensorInfo *input, + const int block_shape_x, + const int block_shape_y, + const Size2D &padding_left, + const Size2D &padding_right, + const ITensorInfo *output); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/functions/CLSpaceToDepthLayer.h b/arm_compute/runtime/CL/functions/CLSpaceToDepthLayer.h index 8a47e95f9d..1b0dfc2b74 100644 --- a/arm_compute/runtime/CL/functions/CLSpaceToDepthLayer.h +++ b/arm_compute/runtime/CL/functions/CLSpaceToDepthLayer.h @@ -75,7 +75,8 @@ public: * @param[out] output Tensor output. Data types supported: same as @p input * @param[in] block_shape Block shape value. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t block_shape); + void + configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, int32_t block_shape); /** Static function to check if given info will lead to a valid configuration of @ref CLSpaceToDepthLayer. * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. diff --git a/arm_compute/runtime/CL/functions/CLSplit.h b/arm_compute/runtime/CL/functions/CLSplit.h index 86c7bdde7d..8d13755212 100644 --- a/arm_compute/runtime/CL/functions/CLSplit.h +++ b/arm_compute/runtime/CL/functions/CLSplit.h @@ -26,7 +26,6 @@ #include "arm_compute/core/CL/ICLTensor.h" #include "arm_compute/core/Types.h" - #include "arm_compute/runtime/CL/functions/CLSlice.h" #include "arm_compute/runtime/CPP/functions/CPPSplit.h" #include "arm_compute/runtime/IFunction.h" diff --git a/arm_compute/runtime/CL/functions/CLStackLayer.h b/arm_compute/runtime/CL/functions/CLStackLayer.h index 54c903a706..18745c8a4f 100644 --- a/arm_compute/runtime/CL/functions/CLStackLayer.h +++ b/arm_compute/runtime/CL/functions/CLStackLayer.h @@ -85,7 +85,10 @@ public: * Negative values wrap around * @param[out] output Output tensor. Data types supported: Same as @p input. */ - void configure(const CLCompileContext &compile_context, const std::vector<ICLTensor *> &input, int axis, ICLTensor *output); + void configure(const CLCompileContext &compile_context, + const std::vector<ICLTensor *> &input, + int axis, + ICLTensor *output); /** Static function to check if given info will lead to a valid configuration of @ref CLStackLayerKernel * * @note Supported input tensor rank: up to 4 diff --git a/arm_compute/runtime/CL/functions/CLStridedSlice.h b/arm_compute/runtime/CL/functions/CLStridedSlice.h index 6fab0c0186..b1edc2481c 100644 --- a/arm_compute/runtime/CL/functions/CLStridedSlice.h +++ b/arm_compute/runtime/CL/functions/CLStridedSlice.h @@ -74,9 +74,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - void configure(const ICLTensor *input, ICLTensor *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + void configure(const ICLTensor *input, + ICLTensor *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Configure kernel * * @note Supported tensor rank: up to 4 @@ -92,9 +97,15 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Static function to check if given info will lead to a valid configuration of @ref CLStridedSlice * @@ -110,9 +121,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); // Inherited methods overridden: void run() override; @@ -143,9 +159,15 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - void configure(const CLCompileContext &compile_context, const ITensorInfo *input, ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + void configure(const CLCompileContext &compile_context, + const ITensorInfo *input, + ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Static function to check if given info will lead to a valid configuration of @ref CLStridedSlice * @@ -161,9 +183,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); }; } // namespace experimental } // namespace arm_compute diff --git a/arm_compute/runtime/CL/functions/CLTile.h b/arm_compute/runtime/CL/functions/CLTile.h index c266adbbd4..4c414670a5 100644 --- a/arm_compute/runtime/CL/functions/CLTile.h +++ b/arm_compute/runtime/CL/functions/CLTile.h @@ -59,7 +59,10 @@ public: * @param[in] multiples Contains the number of times the input tensor should be replicated on the given dimension. * @param[out] output Destination tensor. Same as @p input */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, ICLTensor *output, const Multiples &multiples); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + ICLTensor *output, + const Multiples &multiples); /** Static function to check if given info will lead to a valid configuration of @ref CLTile * * @param[in] input Source tensor info. Data type supported: All. diff --git a/arm_compute/runtime/CL/functions/CLTranspose.h b/arm_compute/runtime/CL/functions/CLTranspose.h index a866aeabaa..9dc977fbeb 100644 --- a/arm_compute/runtime/CL/functions/CLTranspose.h +++ b/arm_compute/runtime/CL/functions/CLTranspose.h @@ -88,6 +88,6 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CLTRANSPOSE_H */ diff --git a/arm_compute/runtime/CL/functions/CLUnstack.h b/arm_compute/runtime/CL/functions/CLUnstack.h index 32ad439b70..a6eee43177 100644 --- a/arm_compute/runtime/CL/functions/CLUnstack.h +++ b/arm_compute/runtime/CL/functions/CLUnstack.h @@ -26,9 +26,8 @@ #include "arm_compute/core/CL/OpenCL.h" #include "arm_compute/core/Types.h" -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/CL/functions/CLStridedSlice.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -72,7 +71,10 @@ public: * @param[in] axis The axis to unstack along. Valid values are [-R,R) where R is the input's rank. Negative values wrap around. * */ - void configure(const CLCompileContext &compile_context, const ICLTensor *input, const std::vector<ICLTensor *> &output_vector, int axis); + void configure(const CLCompileContext &compile_context, + const ICLTensor *input, + const std::vector<ICLTensor *> &output_vector, + int axis); /** Static function to check if given info will lead to a valid configuration of @ref CLUnstack * * @param[in] input Input tensor info. Data type supported: All. diff --git a/arm_compute/runtime/CL/functions/CLWinogradConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLWinogradConvolutionLayer.h index adf5f18626..efea9a1550 100644 --- a/arm_compute/runtime/CL/functions/CLWinogradConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLWinogradConvolutionLayer.h @@ -84,8 +84,13 @@ public: * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation * available which may introduce a drop of accuracy as well. Default is false */ - void configure(ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + void configure(ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); /** Set the input and output tensors. * * @note: This function only works with 3x3,3x1,1x3,5x5,5x1,1x5,7x1 and 1x7 kernels along with unit strides for both NCHW and NHWC data layout @@ -104,8 +109,14 @@ public: * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation * available which may introduce a drop of accuracy as well. Default is false */ - void configure(const CLCompileContext &compile_context, ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + void configure(const CLCompileContext &compile_context, + ICLTensor *input, + const ICLTensor *weights, + const ICLTensor *biases, + ICLTensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); /** Static function to check if given info will lead to a valid configuration of @ref CLWinogradConvolutionLayer * * @note: This function only works with 3x3,3x1,1x3,5x5,5x1 and 1x5 kernels along with unit strides for both NCHW and NHWC data layout @@ -125,8 +136,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CL/tuners/CLTuningParametersList.h b/arm_compute/runtime/CL/tuners/CLTuningParametersList.h index 69572c98d2..5f6d12b4a7 100644 --- a/arm_compute/runtime/CL/tuners/CLTuningParametersList.h +++ b/arm_compute/runtime/CL/tuners/CLTuningParametersList.h @@ -29,6 +29,7 @@ #include "arm_compute/core/Helpers.h" #include "arm_compute/runtime/CL/CLTunerTypes.h" #include "arm_compute/runtime/CL/CLTuningParams.h" + #include "support/ToolchainSupport.h" #include <memory> diff --git a/arm_compute/runtime/CPP/CPPScheduler.h b/arm_compute/runtime/CPP/CPPScheduler.h index a5932d6301..7f70b5fa1f 100644 --- a/arm_compute/runtime/CPP/CPPScheduler.h +++ b/arm_compute/runtime/CPP/CPPScheduler.h @@ -55,10 +55,10 @@ public: static CPPScheduler &get(); // Inherited functions overridden - void set_num_threads(unsigned int num_threads) override; - void set_num_threads_with_affinity(unsigned int num_threads, BindFunc func) override; + void set_num_threads(unsigned int num_threads) override; + void set_num_threads_with_affinity(unsigned int num_threads, BindFunc func) override; unsigned int num_threads() const override; - void schedule(ICPPKernel *kernel, const Hints &hints) override; + void schedule(ICPPKernel *kernel, const Hints &hints) override; void schedule_op(ICPPKernel *kernel, const Hints &hints, const Window &window, ITensorPack &tensors) override; protected: diff --git a/arm_compute/runtime/CPP/functions/CPPBoxWithNonMaximaSuppressionLimit.h b/arm_compute/runtime/CPP/functions/CPPBoxWithNonMaximaSuppressionLimit.h index 58b4bf25cc..9af4ed6208 100644 --- a/arm_compute/runtime/CPP/functions/CPPBoxWithNonMaximaSuppressionLimit.h +++ b/arm_compute/runtime/CPP/functions/CPPBoxWithNonMaximaSuppressionLimit.h @@ -61,8 +61,16 @@ public: * @param[in] keeps_size (Optional) Number of filtered indices per class tensor of size [num_classes]. Data types supported: U32. * @param[in] info (Optional) BoxNMSLimitInfo information. */ - void configure(const ITensor *scores_in, const ITensor *boxes_in, const ITensor *batch_splits_in, ITensor *scores_out, ITensor *boxes_out, ITensor *classes, - ITensor *batch_splits_out = nullptr, ITensor *keeps = nullptr, ITensor *keeps_size = nullptr, const BoxNMSLimitInfo info = BoxNMSLimitInfo()); + void configure(const ITensor *scores_in, + const ITensor *boxes_in, + const ITensor *batch_splits_in, + ITensor *scores_out, + ITensor *boxes_out, + ITensor *classes, + ITensor *batch_splits_out = nullptr, + ITensor *keeps = nullptr, + ITensor *keeps_size = nullptr, + const BoxNMSLimitInfo info = BoxNMSLimitInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CPPDetectionOutputLayer * * @param[in] scores_in The scores input tensor of size [count, num_classes]. Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32 @@ -81,9 +89,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *scores_in, const ITensorInfo *boxes_in, const ITensorInfo *batch_splits_in, const ITensorInfo *scores_out, const ITensorInfo *boxes_out, - const ITensorInfo *classes, - const ITensorInfo *batch_splits_out = nullptr, const ITensorInfo *keeps = nullptr, const ITensorInfo *keeps_size = nullptr, const BoxNMSLimitInfo info = BoxNMSLimitInfo()); + static Status validate(const ITensorInfo *scores_in, + const ITensorInfo *boxes_in, + const ITensorInfo *batch_splits_in, + const ITensorInfo *scores_out, + const ITensorInfo *boxes_out, + const ITensorInfo *classes, + const ITensorInfo *batch_splits_out = nullptr, + const ITensorInfo *keeps = nullptr, + const ITensorInfo *keeps_size = nullptr, + const BoxNMSLimitInfo info = BoxNMSLimitInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CPP/functions/CPPDetectionOutputLayer.h b/arm_compute/runtime/CPP/functions/CPPDetectionOutputLayer.h index f2c7ccccc5..dc8c8e76ba 100644 --- a/arm_compute/runtime/CPP/functions/CPPDetectionOutputLayer.h +++ b/arm_compute/runtime/CPP/functions/CPPDetectionOutputLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPP_DETECTION_OUTPUT_LAYER_H #define ARM_COMPUTE_CPP_DETECTION_OUTPUT_LAYER_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -52,7 +51,11 @@ public: * * @note Output contains all the detections. Of those, only the ones selected by the valid region are valid. */ - void configure(const ITensor *input_loc, const ITensor *input_conf, const ITensor *input_priorbox, ITensor *output, DetectionOutputLayerInfo info = DetectionOutputLayerInfo()); + void configure(const ITensor *input_loc, + const ITensor *input_conf, + const ITensor *input_priorbox, + ITensor *output, + DetectionOutputLayerInfo info = DetectionOutputLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CPPDetectionOutputLayer * * @param[in] input_loc The mbox location input tensor info. Data types supported: F32. @@ -63,7 +66,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input_loc, const ITensorInfo *input_conf, const ITensorInfo *input_priorbox, const ITensorInfo *output, + static Status validate(const ITensorInfo *input_loc, + const ITensorInfo *input_conf, + const ITensorInfo *input_priorbox, + const ITensorInfo *output, DetectionOutputLayerInfo info = DetectionOutputLayerInfo()); // Inherited methods overridden: void run() override; @@ -82,12 +88,12 @@ private: int _num_priors; int _num; - std::vector<LabelBBox> _all_location_predictions; + std::vector<LabelBBox> _all_location_predictions; std::vector<std::map<int, std::vector<float>>> _all_confidence_scores; - std::vector<BBox> _all_prior_bboxes; - std::vector<std::array<float, 4>> _all_prior_variances; - std::vector<LabelBBox> _all_decode_bboxes; - std::vector<std::map<int, std::vector<int>>> _all_indices; + std::vector<BBox> _all_prior_bboxes; + std::vector<std::array<float, 4>> _all_prior_variances; + std::vector<LabelBBox> _all_decode_bboxes; + std::vector<std::map<int, std::vector<int>>> _all_indices; }; } // namespace arm_compute #endif /* ARM_COMPUTE_CPP_DETECTION_OUTPUT_LAYER_H */ diff --git a/arm_compute/runtime/CPP/functions/CPPDetectionPostProcessLayer.h b/arm_compute/runtime/CPP/functions/CPPDetectionPostProcessLayer.h index 94248ff314..a40e4f9ecb 100644 --- a/arm_compute/runtime/CPP/functions/CPPDetectionPostProcessLayer.h +++ b/arm_compute/runtime/CPP/functions/CPPDetectionPostProcessLayer.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_CPP_DETECTION_POSTPROCESS_H #define ARM_COMPUTE_CPP_DETECTION_POSTPROCESS_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/runtime/CPP/functions/CPPNonMaximumSuppression.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/Tensor.h" @@ -65,8 +64,14 @@ public: * * @note Output contains all the detections. Of those, only the ones selected by the valid region are valid. */ - void configure(const ITensor *input_box_encoding, const ITensor *input_score, const ITensor *input_anchors, - ITensor *output_boxes, ITensor *output_classes, ITensor *output_scores, ITensor *num_detection, DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); + void configure(const ITensor *input_box_encoding, + const ITensor *input_score, + const ITensor *input_anchors, + ITensor *output_boxes, + ITensor *output_classes, + ITensor *output_scores, + ITensor *num_detection, + DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref CPPDetectionPostProcessLayer * * @param[in] input_box_encoding The bounding box input tensor info. Data types supported: F32/QASYMM8/QASYMM8_SIGNED. @@ -80,8 +85,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input_box_encoding, const ITensorInfo *input_class_score, const ITensorInfo *input_anchors, - ITensorInfo *output_boxes, ITensorInfo *output_classes, ITensorInfo *output_scores, ITensorInfo *num_detection, + static Status validate(const ITensorInfo *input_box_encoding, + const ITensorInfo *input_class_score, + const ITensorInfo *input_anchors, + ITensorInfo *output_boxes, + ITensorInfo *output_classes, + ITensorInfo *output_scores, + ITensorInfo *num_detection, DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/CPP/functions/CPPNonMaximumSuppression.h b/arm_compute/runtime/CPP/functions/CPPNonMaximumSuppression.h index 71c44a8bd1..af6afc6029 100644 --- a/arm_compute/runtime/CPP/functions/CPPNonMaximumSuppression.h +++ b/arm_compute/runtime/CPP/functions/CPPNonMaximumSuppression.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPP_NONMAXIMUMSUPPRESSION_LAYER_H #define ARM_COMPUTE_CPP_NONMAXIMUMSUPPRESSION_LAYER_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -48,7 +47,12 @@ public: * @param[in] nms_threshold The threshold used in non maximum suppression. * */ - void configure(const ITensor *bboxes, const ITensor *scores, ITensor *indices, unsigned int max_output_size, const float score_threshold, const float nms_threshold); + void configure(const ITensor *bboxes, + const ITensor *scores, + ITensor *indices, + unsigned int max_output_size, + const float score_threshold, + const float nms_threshold); /** Static function to check if given arguments will lead to a valid configuration of @ref CPPNonMaximumSuppression * @@ -60,8 +64,12 @@ public: * @param[in] nms_threshold The threshold used in non maximum suppression. * */ - static Status validate(const ITensorInfo *bboxes, const ITensorInfo *scores, const ITensorInfo *indices, unsigned int max_output_size, - const float score_threshold, const float nms_threshold); + static Status validate(const ITensorInfo *bboxes, + const ITensorInfo *scores, + const ITensorInfo *indices, + unsigned int max_output_size, + const float score_threshold, + const float nms_threshold); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CPP_NONMAXIMUMSUPPRESSION_LAYER_H */ diff --git a/arm_compute/runtime/CPP/functions/CPPPermute.h b/arm_compute/runtime/CPP/functions/CPPPermute.h index 85c1502324..232da41b8e 100644 --- a/arm_compute/runtime/CPP/functions/CPPPermute.h +++ b/arm_compute/runtime/CPP/functions/CPPPermute.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPPPERMUTE_H #define ARM_COMPUTE_CPPPERMUTE_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -53,5 +52,5 @@ public: */ static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PermutationVector &perm); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CPPPERMUTE_H */ diff --git a/arm_compute/runtime/CPP/functions/CPPSplit.h b/arm_compute/runtime/CPP/functions/CPPSplit.h index 56aad2db4b..9be081f5bb 100644 --- a/arm_compute/runtime/CPP/functions/CPPSplit.h +++ b/arm_compute/runtime/CPP/functions/CPPSplit.h @@ -29,7 +29,6 @@ #include "arm_compute/core/TensorInfo.h" #include "arm_compute/core/Types.h" #include "arm_compute/core/utils/misc/ShapeCalculator.h" - #include "arm_compute/runtime/IFunction.h" namespace arm_compute @@ -39,8 +38,7 @@ template <typename SliceType, typename TensorInterfaceType = ITensor> class CPPSplit : public IFunction { public: - CPPSplit() - : _outputs_vector(), _slice_functions(), _num_outputs(0) + CPPSplit() : _outputs_vector(), _slice_functions(), _num_outputs(0) { } /** Static function to check if given info will lead to a valid configuration of @ref CPPSplit @@ -64,14 +62,16 @@ public: unsigned int total_output_shape_size = 0; // Sum the output sizes and fall back to evenly-sized splits if any are zero - const bool using_split_shapes = std::none_of(outputs.begin(), outputs.end(), [&total_output_shape_size](ITensorInfo * info) - { - unsigned int output_shape_size = info->tensor_shape().total_size(); - total_output_shape_size += output_shape_size; - return output_shape_size == 0; - }); - - if(using_split_shapes) + const bool using_split_shapes = std::none_of(outputs.begin(), outputs.end(), + [&total_output_shape_size](ITensorInfo *info) + { + unsigned int output_shape_size = + info->tensor_shape().total_size(); + total_output_shape_size += output_shape_size; + return output_shape_size == 0; + }); + + if (using_split_shapes) { ARM_COMPUTE_RETURN_ERROR_ON(input->tensor_shape().total_size() != total_output_shape_size); } @@ -83,10 +83,10 @@ public: // Validate output tensors unsigned int axis_offset = 0; - for(const auto &output : outputs) + for (const auto &output : outputs) { ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(output); - if(using_split_shapes) + if (using_split_shapes) { output_shape = output->tensor_shape(); ARM_COMPUTE_RETURN_ERROR_ON(output_shape.total_size() == 0); @@ -97,14 +97,14 @@ public: // Start/End coordinates Coordinates start_coords; Coordinates end_coords; - for(unsigned int d = 0; d < output_shape.num_dimensions(); ++d) + for (unsigned int d = 0; d < output_shape.num_dimensions(); ++d) { end_coords.set(d, -1); } // Output auto inizialitation if not yet initialized TensorInfo tmp_output_info = *output->clone(); - if(tmp_output_info.tensor_shape().total_size() == 0) + if (tmp_output_info.tensor_shape().total_size() == 0) { tmp_output_info = input->clone()->set_is_resizable(true).set_tensor_shape(output_shape); } @@ -128,7 +128,8 @@ public: * from the split dimension. * @param[in] axis Axis on which to split the input. */ - void configure(const TensorInterfaceType *input, const std::vector<TensorInterfaceType *> &outputs, unsigned int axis) + void + configure(const TensorInterfaceType *input, const std::vector<TensorInterfaceType *> &outputs, unsigned int axis) { // Create Slice functions _num_outputs = outputs.size(); @@ -136,17 +137,16 @@ public: // Extract output tensor info std::vector<ITensorInfo *> outputs_info; - for(auto &output : outputs) + for (auto &output : outputs) { ARM_COMPUTE_ERROR_ON_NULLPTR(output); outputs_info.emplace_back(output->info()); } // If any of the outputs have a zero size, fall-back to using evenly-sized output splits - const bool outputs_have_sizes = std::none_of(outputs_info.begin(), outputs_info.end(), [](ITensorInfo * info) - { - return info->tensor_shape().total_size() == 0; - }); + const bool outputs_have_sizes = + std::none_of(outputs_info.begin(), outputs_info.end(), + [](ITensorInfo *info) { return info->tensor_shape().total_size() == 0; }); // Validate ARM_COMPUTE_ERROR_THROW_ON(CPPSplit::validate(input->info(), outputs_info, axis)); @@ -154,12 +154,13 @@ public: unsigned int axis_offset = 0; unsigned int i = 0; - for(const auto &output_info : outputs_info) + for (const auto &output_info : outputs_info) { // Get output shape - TensorShape output_shape = (outputs_have_sizes ? - output_info->tensor_shape() : - arm_compute::misc::shape_calculator::compute_split_shape(input->info(), axis, _num_outputs)); + TensorShape output_shape = + (outputs_have_sizes + ? output_info->tensor_shape() + : arm_compute::misc::shape_calculator::compute_split_shape(input->info(), axis, _num_outputs)); const size_t axis_split_step = output_shape[axis]; @@ -167,7 +168,7 @@ public: Coordinates start_coords; Coordinates end_coords; - for(unsigned int d = 0; d < output_shape.num_dimensions(); ++d) + for (unsigned int d = 0; d < output_shape.num_dimensions(); ++d) { end_coords.set(d, -1); } diff --git a/arm_compute/runtime/CPP/functions/CPPTopKV.h b/arm_compute/runtime/CPP/functions/CPPTopKV.h index 2f63084056..232cbb3067 100644 --- a/arm_compute/runtime/CPP/functions/CPPTopKV.h +++ b/arm_compute/runtime/CPP/functions/CPPTopKV.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPPTOPKV_H #define ARM_COMPUTE_CPPTOPKV_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -54,7 +53,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *predictions, const ITensorInfo *targets, ITensorInfo *output, const unsigned int k); + static Status + validate(const ITensorInfo *predictions, const ITensorInfo *targets, ITensorInfo *output, const unsigned int k); }; } // namespace arm_compute #endif /* ARM_COMPUTE_CPPTOPKV_H */ diff --git a/arm_compute/runtime/CPP/functions/CPPUpsample.h b/arm_compute/runtime/CPP/functions/CPPUpsample.h index b97d4d1cc1..3b0f997b17 100644 --- a/arm_compute/runtime/CPP/functions/CPPUpsample.h +++ b/arm_compute/runtime/CPP/functions/CPPUpsample.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_CPPUPSAMPLE_H #define ARM_COMPUTE_CPPUPSAMPLE_H -#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/ICPPSimpleFunction.h" namespace arm_compute { @@ -44,5 +43,5 @@ public: */ void configure(const ITensor *input, ITensor *output, const PadStrideInfo &info); }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_CPPUPSAMPLE_H */ diff --git a/arm_compute/runtime/FunctionDescriptors.h b/arm_compute/runtime/FunctionDescriptors.h index 05f172b9f1..c3af17d6f2 100644 --- a/arm_compute/runtime/FunctionDescriptors.h +++ b/arm_compute/runtime/FunctionDescriptors.h @@ -41,16 +41,16 @@ enum class FFTDirection /** Descriptor used by the FFT1D function */ struct FFT1DInfo { - unsigned int axis{ 0 }; /**< Axis to run the FFT on. */ - FFTDirection direction{ FFTDirection::Forward }; /**< Direction of the FFT. */ + unsigned int axis{0}; /**< Axis to run the FFT on. */ + FFTDirection direction{FFTDirection::Forward}; /**< Direction of the FFT. */ }; /** Descriptor used by the FFT2D function */ struct FFT2DInfo { - unsigned int axis0{ 0 }; /**< Axis to run first pass on. If same, multiple transforms are performed on single axis*/ - unsigned int axis1{ 1 }; /**< Axis to run second pass on. If same, multiple transforms are performed on single axis*/ - FFTDirection direction{ FFTDirection::Forward }; /**< Direction of the FFT. */ + unsigned int axis0{0}; /**< Axis to run first pass on. If same, multiple transforms are performed on single axis*/ + unsigned int axis1{1}; /**< Axis to run second pass on. If same, multiple transforms are performed on single axis*/ + FFTDirection direction{FFTDirection::Forward}; /**< Direction of the FFT. */ }; /** Descriptor used by the 2d Convolution function */ @@ -64,15 +64,20 @@ struct Conv2dInfo bool enable_fast_math, unsigned int num_groups, const WeightsInfo &weights_info = WeightsInfo()) - : conv_info(conv_info), dilation(dilation), act_info(act_info), enable_fast_math(enable_fast_math), num_groups(num_groups), weights_info(weights_info) + : conv_info(conv_info), + dilation(dilation), + act_info(act_info), + enable_fast_math(enable_fast_math), + num_groups(num_groups), + weights_info(weights_info) { } PadStrideInfo conv_info{}; - Size2D dilation{ 1U, 1U }; + Size2D dilation{1U, 1U}; ActivationLayerInfo act_info{}; - bool enable_fast_math{ false }; - unsigned int num_groups{ 1 }; + bool enable_fast_math{false}; + unsigned int num_groups{1}; WeightsInfo weights_info{}; }; @@ -87,16 +92,21 @@ struct Conv3dInfo const Size3D &dilation, const DimensionRoundingType &round_type, bool enable_fast_math) - : stride(stride), padding(padding), act_info(act_info), dilation(dilation), round_type(round_type), enable_fast_math(enable_fast_math) + : stride(stride), + padding(padding), + act_info(act_info), + dilation(dilation), + round_type(round_type), + enable_fast_math(enable_fast_math) { } - Size3D stride{ 1U, 1U, 1U }; + Size3D stride{1U, 1U, 1U}; Padding3D padding{}; ActivationLayerInfo act_info{}; - Size3D dilation{ 1U, 1U, 1U }; + Size3D dilation{1U, 1U, 1U}; DimensionRoundingType round_type{}; - bool enable_fast_math{ false }; + bool enable_fast_math{false}; }; } // namespace arm_compute diff --git a/arm_compute/runtime/IAllocator.h b/arm_compute/runtime/IAllocator.h index 5c28b24fea..f8446db811 100644 --- a/arm_compute/runtime/IAllocator.h +++ b/arm_compute/runtime/IAllocator.h @@ -56,5 +56,5 @@ public: */ virtual std::unique_ptr<IMemoryRegion> make_region(size_t size, size_t alignment) = 0; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_IALLOCATOR_H */ diff --git a/arm_compute/runtime/IFunction.h b/arm_compute/runtime/IFunction.h index b7b28f999d..fb68dbbecf 100644 --- a/arm_compute/runtime/IFunction.h +++ b/arm_compute/runtime/IFunction.h @@ -58,5 +58,5 @@ public: { } }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_IFUNCTION_H */ diff --git a/arm_compute/runtime/IMemoryGroup.h b/arm_compute/runtime/IMemoryGroup.h index a977a4a3c3..77198dd29d 100644 --- a/arm_compute/runtime/IMemoryGroup.h +++ b/arm_compute/runtime/IMemoryGroup.h @@ -86,8 +86,7 @@ public: * * @param[in] memory_group Memory group to handle */ - explicit MemoryGroupResourceScope(IMemoryGroup &memory_group) - : _memory_group(memory_group) + explicit MemoryGroupResourceScope(IMemoryGroup &memory_group) : _memory_group(memory_group) { _memory_group.acquire(); } @@ -100,5 +99,5 @@ public: private: IMemoryGroup &_memory_group; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_IMEMORYGROUP_H */ diff --git a/arm_compute/runtime/IMemoryManager.h b/arm_compute/runtime/IMemoryManager.h index 4d7d8cd9c9..42910edfda 100644 --- a/arm_compute/runtime/IMemoryManager.h +++ b/arm_compute/runtime/IMemoryManager.h @@ -65,5 +65,5 @@ public: */ virtual void clear() = 0; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_IMEMORYMANAGER_H */ diff --git a/arm_compute/runtime/IMemoryPool.h b/arm_compute/runtime/IMemoryPool.h index b8d36c362d..0c112c8f35 100644 --- a/arm_compute/runtime/IMemoryPool.h +++ b/arm_compute/runtime/IMemoryPool.h @@ -60,5 +60,5 @@ public: */ virtual std::unique_ptr<IMemoryPool> duplicate() = 0; }; -} // arm_compute +} // namespace arm_compute #endif /* ARM_COMPUTE_IMEMORYPOOL_H */ diff --git a/arm_compute/runtime/IMemoryRegion.h b/arm_compute/runtime/IMemoryRegion.h index 914aa57fbe..9431663e4e 100644 --- a/arm_compute/runtime/IMemoryRegion.h +++ b/arm_compute/runtime/IMemoryRegion.h @@ -37,8 +37,7 @@ public: * * @param[in] size Region size */ - explicit IMemoryRegion(size_t size) - : _size(size) + explicit IMemoryRegion(size_t size) : _size(size) { } /** Virtual Destructor */ diff --git a/arm_compute/runtime/IPoolManager.h b/arm_compute/runtime/IPoolManager.h index 481bde5fb6..5f6d4ffbe5 100644 --- a/arm_compute/runtime/IPoolManager.h +++ b/arm_compute/runtime/IPoolManager.h @@ -69,5 +69,5 @@ public: */ virtual size_t num_pools() const = 0; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_IPOOLMANAGER_H */ diff --git a/arm_compute/runtime/IScheduler.h b/arm_compute/runtime/IScheduler.h index df5a44001f..ae204c8560 100644 --- a/arm_compute/runtime/IScheduler.h +++ b/arm_compute/runtime/IScheduler.h @@ -25,8 +25,8 @@ #define ARM_COMPUTE_ISCHEDULER_H #include "arm_compute/core/CPP/CPPTypes.h" -#include "arm_compute/core/Types.h" #include "arm_compute/core/experimental/Types.h" +#include "arm_compute/core/Types.h" #include <functional> #include <limits> @@ -226,7 +226,11 @@ protected: * * @return Adjusted number of windows */ - std::size_t adjust_num_of_windows(const Window &window, std::size_t split_dimension, std::size_t init_num_windows, const ICPPKernel &kernel, const CPUInfo &cpu_info); + std::size_t adjust_num_of_windows(const Window &window, + std::size_t split_dimension, + std::size_t init_num_windows, + const ICPPKernel &kernel, + const CPUInfo &cpu_info); private: unsigned int _num_threads_hint = {}; diff --git a/arm_compute/runtime/ISimpleLifetimeManager.h b/arm_compute/runtime/ISimpleLifetimeManager.h index b2d17c6fea..9e481bb563 100644 --- a/arm_compute/runtime/ISimpleLifetimeManager.h +++ b/arm_compute/runtime/ISimpleLifetimeManager.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_ISIMPLELIFETIMEMANAGER_H #include "arm_compute/runtime/ILifetimeManager.h" - #include "arm_compute/runtime/IMemoryPool.h" #include "arm_compute/runtime/Types.h" @@ -70,7 +69,11 @@ protected: /** Element struct */ struct Element { - Element(void *id_ = nullptr, IMemory *handle_ = nullptr, size_t size_ = 0, size_t alignment_ = 0, bool status_ = false) + Element(void *id_ = nullptr, + IMemory *handle_ = nullptr, + size_t size_ = 0, + size_t alignment_ = 0, + bool status_ = false) : id(id_), handle(handle_), size(size_), alignment(alignment_), status(status_) { } @@ -90,11 +93,12 @@ protected: std::set<void *> bound_elements; }; - IMemoryGroup *_active_group; /**< Active group */ - std::map<void *, Element> _active_elements; /**< A map that contains the active elements */ - std::list<Blob> _free_blobs; /**< Free blobs */ - std::list<Blob> _occupied_blobs; /**< Occupied blobs */ - std::map<IMemoryGroup *, std::map<void *, Element>> _finalized_groups; /**< A map that contains the finalized groups */ + IMemoryGroup *_active_group; /**< Active group */ + std::map<void *, Element> _active_elements; /**< A map that contains the active elements */ + std::list<Blob> _free_blobs; /**< Free blobs */ + std::list<Blob> _occupied_blobs; /**< Occupied blobs */ + std::map<IMemoryGroup *, std::map<void *, Element>> + _finalized_groups; /**< A map that contains the finalized groups */ }; } // namespace arm_compute #endif /* ARM_COMPUTE_ISIMPLELIFETIMEMANAGER_H */ diff --git a/arm_compute/runtime/ITensorAllocator.h b/arm_compute/runtime/ITensorAllocator.h index 17e581b40e..e2d3536169 100644 --- a/arm_compute/runtime/ITensorAllocator.h +++ b/arm_compute/runtime/ITensorAllocator.h @@ -101,9 +101,9 @@ protected: virtual void unlock() = 0; private: - TensorInfo _info_owned{}; /**< Tensor's metadata. */ - TensorInfo *_info_external{ nullptr }; /**< External Tensor's metadata */ - size_t _alignment{}; /**< Tensor's alignment in bytes */ + TensorInfo _info_owned{}; /**< Tensor's metadata. */ + TensorInfo *_info_external{nullptr}; /**< External Tensor's metadata */ + size_t _alignment{}; /**< Tensor's alignment in bytes */ }; } // namespace arm_compute #endif /*ARM_COMPUTE_ITENSORALLOCATOR_H */ diff --git a/arm_compute/runtime/ITransformWeights.h b/arm_compute/runtime/ITransformWeights.h index f85b7966c5..08671bbe3c 100644 --- a/arm_compute/runtime/ITransformWeights.h +++ b/arm_compute/runtime/ITransformWeights.h @@ -72,7 +72,7 @@ public: /** Allow instances of this class to be moved */ ITransformWeights &operator=(ITransformWeights &&other) { - if(this != &other) + if (this != &other) { _num_refcount = other._num_refcount.load(); _reshape_run = other._reshape_run; @@ -119,9 +119,9 @@ public: } protected: - std::atomic<int32_t> _num_refcount{ 0 }; - bool _reshape_run{ false }; + std::atomic<int32_t> _num_refcount{0}; + bool _reshape_run{false}; }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_ITRANSFORMWEIGHTS_H */ diff --git a/arm_compute/runtime/IWeightsManager.h b/arm_compute/runtime/IWeightsManager.h index 3b97d696bb..de8a92faa3 100644 --- a/arm_compute/runtime/IWeightsManager.h +++ b/arm_compute/runtime/IWeightsManager.h @@ -90,8 +90,8 @@ public: private: struct CounterElement { - bool is_unused{ false }; - std::atomic<int> counter{ 1 }; + bool is_unused{false}; + std::atomic<int> counter{1}; }; private: @@ -99,5 +99,5 @@ private: std::map<const ITensor *, CounterElement> _managed_counter; std::map<const ITensor *, ITransformWeights *> _managed_weights_parents; }; -} // arm_compute -#endif /*ARM_COMPUTE_IWEIGHTSMANAGER_H */
\ No newline at end of file +} // namespace arm_compute +#endif /*ARM_COMPUTE_IWEIGHTSMANAGER_H */ diff --git a/arm_compute/runtime/Memory.h b/arm_compute/runtime/Memory.h index 1eab605d50..63514c409b 100644 --- a/arm_compute/runtime/Memory.h +++ b/arm_compute/runtime/Memory.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_MEMORY_H #include "arm_compute/runtime/IMemory.h" - #include "arm_compute/runtime/IMemoryRegion.h" #include <cstddef> @@ -64,8 +63,8 @@ public: // Inherited methods overridden: IMemoryRegion *region() final; IMemoryRegion *region() const final; - void set_region(IMemoryRegion *region) final; - void set_owned_region(std::unique_ptr<IMemoryRegion> region) final; + void set_region(IMemoryRegion *region) final; + void set_owned_region(std::unique_ptr<IMemoryRegion> region) final; private: IMemoryRegion *_region; diff --git a/arm_compute/runtime/MemoryGroup.h b/arm_compute/runtime/MemoryGroup.h index 9fd2b9fa72..93ea3d2c72 100644 --- a/arm_compute/runtime/MemoryGroup.h +++ b/arm_compute/runtime/MemoryGroup.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_MEMORYGROUP_H #define ARM_COMPUTE_MEMORYGROUP_H -#include "arm_compute/runtime/IMemoryGroup.h" - #include "arm_compute/core/Error.h" #include "arm_compute/core/utils/misc/Macros.h" +#include "arm_compute/runtime/IMemoryGroup.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/IMemoryPool.h" @@ -59,8 +58,8 @@ public: // Inherited methods overridden: void manage(IMemoryManageable *obj) override; void finalize_memory(IMemoryManageable *obj, IMemory &obj_memory, size_t size, size_t alignment) override; - void acquire() override; - void release() override; + void acquire() override; + void release() override; MemoryMappings &mappings() override; private: @@ -70,15 +69,13 @@ private: }; inline MemoryGroup::MemoryGroup(std::shared_ptr<IMemoryManager> memory_manager) noexcept - : _memory_manager(memory_manager), - _pool(nullptr), - _mappings() + : _memory_manager(memory_manager), _pool(nullptr), _mappings() { } inline void MemoryGroup::manage(IMemoryManageable *obj) { - if(_memory_manager && (obj != nullptr)) + if (_memory_manager && (obj != nullptr)) { ARM_COMPUTE_ERROR_ON(!_memory_manager->lifetime_manager()); @@ -95,7 +92,7 @@ inline void MemoryGroup::manage(IMemoryManageable *obj) inline void MemoryGroup::finalize_memory(IMemoryManageable *obj, IMemory &obj_memory, size_t size, size_t alignment) { - if(_memory_manager) + if (_memory_manager) { ARM_COMPUTE_ERROR_ON(!_memory_manager->lifetime_manager()); _memory_manager->lifetime_manager()->end_lifetime(obj, obj_memory, size, alignment); @@ -104,7 +101,7 @@ inline void MemoryGroup::finalize_memory(IMemoryManageable *obj, IMemory &obj_me inline void MemoryGroup::acquire() { - if(!_mappings.empty()) + if (!_mappings.empty()) { ARM_COMPUTE_ERROR_ON(!_memory_manager->pool_manager()); _pool = _memory_manager->pool_manager()->lock_pool(); @@ -114,7 +111,7 @@ inline void MemoryGroup::acquire() inline void MemoryGroup::release() { - if(_pool != nullptr) + if (_pool != nullptr) { ARM_COMPUTE_ERROR_ON(!_memory_manager->pool_manager()); ARM_COMPUTE_ERROR_ON(_mappings.empty()); @@ -128,5 +125,5 @@ inline MemoryMappings &MemoryGroup::mappings() { return _mappings; } -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_MEMORYGROUP_H */ diff --git a/arm_compute/runtime/MemoryManagerOnDemand.h b/arm_compute/runtime/MemoryManagerOnDemand.h index 50547ac38e..7c31fe7f5a 100644 --- a/arm_compute/runtime/MemoryManagerOnDemand.h +++ b/arm_compute/runtime/MemoryManagerOnDemand.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_MEMORY_MANAGER_ON_DEMAND_H #define ARM_COMPUTE_MEMORY_MANAGER_ON_DEMAND_H -#include "arm_compute/runtime/IMemoryManager.h" - #include "arm_compute/runtime/ILifetimeManager.h" #include "arm_compute/runtime/IMemoryGroup.h" +#include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/IPoolManager.h" #include <memory> @@ -39,7 +38,8 @@ class MemoryManagerOnDemand : public IMemoryManager { public: /** Default Constructor */ - MemoryManagerOnDemand(std::shared_ptr<ILifetimeManager> lifetime_manager, std::shared_ptr<IPoolManager> pool_manager); + MemoryManagerOnDemand(std::shared_ptr<ILifetimeManager> lifetime_manager, + std::shared_ptr<IPoolManager> pool_manager); /** Prevent instances of this class to be copy constructed */ MemoryManagerOnDemand(const MemoryManagerOnDemand &) = delete; /** Prevent instances of this class to be copied */ @@ -52,12 +52,12 @@ public: // Inherited methods overridden: ILifetimeManager *lifetime_manager() override; IPoolManager *pool_manager() override; - void populate(IAllocator &allocator, size_t num_pools) override; - void clear() override; + void populate(IAllocator &allocator, size_t num_pools) override; + void clear() override; private: std::shared_ptr<ILifetimeManager> _lifetime_mgr; /**< Lifetime manager */ std::shared_ptr<IPoolManager> _pool_mgr; /**< Memory pool manager */ }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_MEMORY_MANAGER_ON_DEMAND_H */ diff --git a/arm_compute/runtime/MemoryRegion.h b/arm_compute/runtime/MemoryRegion.h index 6408deceaa..f8a4898281 100644 --- a/arm_compute/runtime/MemoryRegion.h +++ b/arm_compute/runtime/MemoryRegion.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_RUNTIME_MEMORY_REGION_H #define ARM_COMPUTE_RUNTIME_MEMORY_REGION_H -#include "arm_compute/runtime/IMemoryRegion.h" - #include "arm_compute/core/Error.h" +#include "arm_compute/runtime/IMemoryRegion.h" #include <cstddef> @@ -41,21 +40,17 @@ public: * @param[in] size Region size * @param[in] alignment Alignment in bytes of the base pointer. Defaults to 0 */ - MemoryRegion(size_t size, size_t alignment = 0) - : IMemoryRegion(size), _mem(nullptr), _ptr(nullptr) + MemoryRegion(size_t size, size_t alignment = 0) : IMemoryRegion(size), _mem(nullptr), _ptr(nullptr) { - if(size != 0) + if (size != 0) { // Allocate backing memory size_t space = size + alignment; - _mem = std::shared_ptr<uint8_t>(new uint8_t[space](), [](uint8_t *ptr) - { - delete[] ptr; - }); - _ptr = _mem.get(); + _mem = std::shared_ptr<uint8_t>(new uint8_t[space](), [](uint8_t *ptr) { delete[] ptr; }); + _ptr = _mem.get(); // Calculate alignment offset - if(alignment != 0) + if (alignment != 0) { void *aligned_ptr = _mem.get(); std::align(alignment, size, aligned_ptr, space); @@ -63,10 +58,9 @@ public: } } } - MemoryRegion(void *ptr, size_t size) - : IMemoryRegion(size), _mem(nullptr), _ptr(nullptr) + MemoryRegion(void *ptr, size_t size) : IMemoryRegion(size), _mem(nullptr), _ptr(nullptr) { - if(size != 0) + if (size != 0) { _ptr = ptr; } @@ -91,7 +85,7 @@ public: } std::unique_ptr<IMemoryRegion> extract_subregion(size_t offset, size_t size) final { - if(_ptr != nullptr && (offset < _size) && (_size - offset >= size)) + if (_ptr != nullptr && (offset < _size) && (_size - offset >= size)) { return std::make_unique<MemoryRegion>(static_cast<uint8_t *>(_ptr) + offset, size); } diff --git a/arm_compute/runtime/NEON/INEOperator.h b/arm_compute/runtime/NEON/INEOperator.h index 5637d831a3..7971168d24 100644 --- a/arm_compute/runtime/NEON/INEOperator.h +++ b/arm_compute/runtime/NEON/INEOperator.h @@ -24,11 +24,11 @@ #ifndef ARM_COMPUTE_INEOPERATOR_H #define ARM_COMPUTE_INEOPERATOR_H -#include "../../core/ITensor.h" #include "arm_compute/runtime/IOperator.h" #include "arm_compute/runtime/IRuntimeContext.h" #include "arm_compute/runtime/Types.h" +#include "../../core/ITensor.h" #include <memory> namespace arm_compute @@ -60,8 +60,8 @@ public: ~INEOperator(); // Inherited methods overridden: - void run(ITensorPack &tensors) override; - void prepare(ITensorPack &constants) override; + void run(ITensorPack &tensors) override; + void prepare(ITensorPack &constants) override; MemoryRequirements workspace() const override; protected: diff --git a/arm_compute/runtime/NEON/INESimpleFunction.h b/arm_compute/runtime/NEON/INESimpleFunction.h index 7512759bd0..f783a836ee 100644 --- a/arm_compute/runtime/NEON/INESimpleFunction.h +++ b/arm_compute/runtime/NEON/INESimpleFunction.h @@ -57,5 +57,5 @@ protected: std::unique_ptr<INEKernel> _kernel; /**< Kernel to run */ std::unique_ptr<NEFillBorderKernel> _border_handler; /**< Kernel to handle image borders */ }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_INESIMPLEFUNCTION_H */ diff --git a/arm_compute/runtime/NEON/NEScheduler.h b/arm_compute/runtime/NEON/NEScheduler.h index a3082d00f6..613f44cc52 100644 --- a/arm_compute/runtime/NEON/NEScheduler.h +++ b/arm_compute/runtime/NEON/NEScheduler.h @@ -30,5 +30,5 @@ namespace arm_compute { /** CPU Scheduler */ using NEScheduler = Scheduler; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_NESCHEDULER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEActivationLayer.h b/arm_compute/runtime/NEON/functions/NEActivationLayer.h index 9992de2af8..5584fdc783 100644 --- a/arm_compute/runtime/NEON/functions/NEActivationLayer.h +++ b/arm_compute/runtime/NEON/functions/NEActivationLayer.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_NEACTIVATIONLAYER_H #define ARM_COMPUTE_NEACTIVATIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IRuntimeContext.h" #include <memory> @@ -102,5 +101,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} // namespace arm_computes +} // namespace arm_compute #endif /* ARM_COMPUTE_NEACTIVATIONLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEAddMulAdd.h b/arm_compute/runtime/NEON/functions/NEAddMulAdd.h index e5e85542f8..6c65c055dd 100644 --- a/arm_compute/runtime/NEON/functions/NEAddMulAdd.h +++ b/arm_compute/runtime/NEON/functions/NEAddMulAdd.h @@ -81,19 +81,28 @@ public: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * */ - void configure(ITensor *input1, ITensor *input2, ITensor *bn_mul, ITensor *bn_add, - ITensor *add_output, ITensor *final_output, - ConvertPolicy policy, const ActivationLayerInfo &act_info); + void configure(ITensor *input1, + ITensor *input2, + ITensor *bn_mul, + ITensor *bn_add, + ITensor *add_output, + ITensor *final_output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info); /** Static function to check if given info will lead to a valid configuration of @ref NEAddMulAdd * * Similar to @ref NEAddMulAdd::configure() except the arguments are @ref ITensorInfo * instead of @ref ITensor * * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, - const ITensorInfo *bn_mul, const ITensorInfo *bn_add, - const ITensorInfo *add_output, const ITensorInfo *final_output, - ConvertPolicy policy, const ActivationLayerInfo &act_info); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *bn_mul, + const ITensorInfo *bn_add, + const ITensorInfo *add_output, + const ITensorInfo *final_output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEArithmeticAddition.h b/arm_compute/runtime/NEON/functions/NEArithmeticAddition.h index b0d710d517..73a43dbc44 100644 --- a/arm_compute/runtime/NEON/functions/NEArithmeticAddition.h +++ b/arm_compute/runtime/NEON/functions/NEArithmeticAddition.h @@ -27,6 +27,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -74,7 +75,11 @@ public: * @param[in] policy Policy to use to handle overflow. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(const ITensor *input1, const ITensor *input2, ITensor *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const ITensor *input1, + const ITensor *input2, + ITensor *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEArithmeticAddition * * @param[in] input1 First tensor input info. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/QSYMM16/F16/S32/F32 @@ -85,7 +90,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h b/arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h index 6fbe9ad450..3e4f6356c5 100644 --- a/arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h +++ b/arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h @@ -80,7 +80,11 @@ public: * @param[in] policy Policy to use to handle overflow. Convert policy cannot be WRAP if datatype is quantized. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(const ITensor *input1, const ITensor *input2, ITensor *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(const ITensor *input1, + const ITensor *input2, + ITensor *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEArithmeticSubtraction * * @param[in] input1 First tensor input. Data types supported: U8/QASYMM8/QASYMM8_SIGNED/S16/S32/F16/F32 @@ -91,7 +95,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ConvertPolicy policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + ConvertPolicy policy, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEBatchNormalizationLayer.h b/arm_compute/runtime/NEON/functions/NEBatchNormalizationLayer.h index ec00fbdbf2..99e2dcadbb 100644 --- a/arm_compute/runtime/NEON/functions/NEBatchNormalizationLayer.h +++ b/arm_compute/runtime/NEON/functions/NEBatchNormalizationLayer.h @@ -81,7 +81,13 @@ public: * @param[in] epsilon (Optional) Small value to avoid division with zero. Default value is 0.001f. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Only RELU, BOUNDED_RELU and LU_BOUNDED_RELU supported. */ - void configure(ITensor *input, ITensor *output, const ITensor *mean, const ITensor *var, const ITensor *beta = nullptr, const ITensor *gamma = nullptr, float epsilon = 0.001f, + void configure(ITensor *input, + ITensor *output, + const ITensor *mean, + const ITensor *var, + const ITensor *beta = nullptr, + const ITensor *gamma = nullptr, + float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEBatchNormalizationLayer * @@ -98,10 +104,14 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const ITensorInfo *mean, const ITensorInfo *var, - const ITensorInfo *beta = nullptr, const ITensorInfo *gamma = nullptr, - float epsilon = 0.001f, ActivationLayerInfo act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const ITensorInfo *mean, + const ITensorInfo *var, + const ITensorInfo *beta = nullptr, + const ITensorInfo *gamma = nullptr, + float epsilon = 0.001f, + ActivationLayerInfo act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -109,5 +119,5 @@ public: private: std::unique_ptr<NEBatchNormalizationLayerKernel> _norm_kernel; /**< Batch normalization layer kernel */ }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEBATCHNORMALIZATIONLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEBatchToSpaceLayer.h b/arm_compute/runtime/NEON/functions/NEBatchToSpaceLayer.h index b33ba435a8..ebed0bea29 100644 --- a/arm_compute/runtime/NEON/functions/NEBatchToSpaceLayer.h +++ b/arm_compute/runtime/NEON/functions/NEBatchToSpaceLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEBATCHTOSPACELAYER_H #define ARM_COMPUTE_NEBATCHTOSPACELAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" namespace arm_compute @@ -77,7 +76,11 @@ public: * @param[out] output Tensor output. Data types supported: same as @p input * @param[in] crop_info Specifies how the output shape is cropped after batch to space is performed */ - void configure(const ITensor *input, int32_t block_shape_x, int32_t block_shape_y, ITensor *output, const CropInfo &crop_info = CropInfo{}); + void configure(const ITensor *input, + int32_t block_shape_x, + int32_t block_shape_y, + ITensor *output, + const CropInfo &crop_info = CropInfo{}); /** Static function to check if given info will lead to a valid configuration of @ref CLBatchToSpaceLayer * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -99,7 +102,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, int32_t block_shape_x, int32_t block_shape_y, const ITensorInfo *output, const CropInfo &crop_info = CropInfo{}); + static Status validate(const ITensorInfo *input, + int32_t block_shape_x, + int32_t block_shape_y, + const ITensorInfo *output, + const CropInfo &crop_info = CropInfo{}); }; } // namespace arm_compute #endif /* ARM_COMPUTE_NEBATCHTOSPACELAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h b/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h index 2a196a2de5..aa41fc0df2 100644 --- a/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h +++ b/arm_compute/runtime/NEON/functions/NEBoundingBoxTransform.h @@ -57,7 +57,8 @@ public: * * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the BoundingBoxTransformInfo struct. */ - void configure(const ITensor *boxes, ITensor *pred_boxes, const ITensor *deltas, const BoundingBoxTransformInfo &info); + void + configure(const ITensor *boxes, ITensor *pred_boxes, const ITensor *deltas, const BoundingBoxTransformInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref NEBoundingBoxTransform * @@ -71,7 +72,10 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *boxes, const ITensorInfo *pred_boxes, const ITensorInfo *deltas, const BoundingBoxTransformInfo &info); + static Status validate(const ITensorInfo *boxes, + const ITensorInfo *pred_boxes, + const ITensorInfo *deltas, + const BoundingBoxTransformInfo &info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_NEBOUNDINGBOXTRANSFORM_H */ diff --git a/arm_compute/runtime/NEON/functions/NECast.h b/arm_compute/runtime/NEON/functions/NECast.h index 821249c142..43cae777f6 100644 --- a/arm_compute/runtime/NEON/functions/NECast.h +++ b/arm_compute/runtime/NEON/functions/NECast.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NECAST_H #define ARM_COMPUTE_NECAST_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h b/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h index dd1c709d76..1600f85488 100644 --- a/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h +++ b/arm_compute/runtime/NEON/functions/NEConcatenateLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NECONCATENATELAYER_H #define ARM_COMPUTE_NECONCATENATELAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -87,7 +86,8 @@ public: * * @return a status */ - static Status validate(const std::vector<const ITensorInfo *> &inputs_vector, const ITensorInfo *output, size_t axis); + static Status + validate(const std::vector<const ITensorInfo *> &inputs_vector, const ITensorInfo *output, size_t axis); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEConv3D.h b/arm_compute/runtime/NEON/functions/NEConv3D.h index 2a3c5351b0..525f37f3e7 100644 --- a/arm_compute/runtime/NEON/functions/NEConv3D.h +++ b/arm_compute/runtime/NEON/functions/NEConv3D.h @@ -24,11 +24,10 @@ #ifndef ARM_COMPUTE_NECONV3D_H #define ARM_COMPUTE_NECONV3D_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/Types.h" #include "arm_compute/runtime/FunctionDescriptors.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -76,14 +75,19 @@ public: * @param[out] output Destination tensor. 4 lower dimensions represent a single output [OFM, width, height, depth], while the rest represent batch of outputs. * @param[in] conv_info Contains padding, stride, acitvation information described in @ref Conv3dInfo. */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const Conv3dInfo &conv_info); + void configure( + ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const Conv3dInfo &conv_info); /** Static function to check if given info will lead to a valid configuration * * Similar to NEConv3D::configure() * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const Conv3dInfo &conv_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const Conv3dInfo &conv_info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h b/arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h index a892d3036b..dc6b22d717 100644 --- a/arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h +++ b/arm_compute/runtime/NEON/functions/NEConvertFullyConnectedWeights.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NECONVERTFULLYCONNECTEDWEIGHTS_H #define ARM_COMPUTE_NECONVERTFULLYCONNECTEDWEIGHTS_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" namespace arm_compute { @@ -66,7 +65,8 @@ public: * @param[in] original_input_shape Shape of the original input tensor (the one entering fully connected layer). * @param[in] data_layout The data layout the weights have been trained in. */ - void configure(const ITensor *input, ITensor *output, const TensorShape &original_input_shape, DataLayout data_layout); + void + configure(const ITensor *input, ITensor *output, const TensorShape &original_input_shape, DataLayout data_layout); /** Static function to check if given info will lead to a valid configuration of @ref NEConvertFullyConnectedWeights * * @param[in] input Source weights tensor info to convert. Must be 2 dimensional. Data types supported: All. @@ -76,7 +76,10 @@ public: * * @return A Status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const TensorShape &original_input_shape, DataLayout data_layout); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const TensorShape &original_input_shape, + DataLayout data_layout); // Inherited methods overriden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEConvolutionLayer.h index 4dd76d082b..cdf0f652e1 100644 --- a/arm_compute/runtime/NEON/functions/NEConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEConvolutionLayer.h @@ -24,11 +24,10 @@ #ifndef ARM_COMPUTE_NECONVOLUTIONLAYER_H #define ARM_COMPUTE_NECONVOLUTIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/ITensorInfo.h" #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" #include <memory> @@ -119,8 +118,16 @@ public: * available which may introduce a drop of accuracy as well. Default is false * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is not supported */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const WeightsInfo &weights_info = WeightsInfo(), - const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, unsigned int num_groups = 1); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if given info will lead to a valid configuration of @ref NEConvolutionLayer * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], @@ -143,9 +150,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, - unsigned int num_groups = 1); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if given info will return the convolution called by @ref NEConvolutionLayer * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], @@ -165,8 +179,14 @@ public: * * @return the Convolution Method Hint */ - static ConvolutionMethod get_convolution_method(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + static ConvolutionMethod get_convolution_method(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; void prepare() override; diff --git a/arm_compute/runtime/NEON/functions/NECopy.h b/arm_compute/runtime/NEON/functions/NECopy.h index ee02c259f4..840c03e968 100644 --- a/arm_compute/runtime/NEON/functions/NECopy.h +++ b/arm_compute/runtime/NEON/functions/NECopy.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NECOPY_H #define ARM_COMPUTE_NECOPY_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NECropResize.h b/arm_compute/runtime/NEON/functions/NECropResize.h index 143bbbc6f1..f806762158 100644 --- a/arm_compute/runtime/NEON/functions/NECropResize.h +++ b/arm_compute/runtime/NEON/functions/NECropResize.h @@ -75,8 +75,13 @@ public: * @param[in] method The policy to be used when resizing image. Default is bilinear. * @param[in] extrapolation_value Value to be used for values outside of the image for cropping and resizing. Default is 0. */ - void configure(const ITensor *input, const ITensor *boxes, const ITensor *box_ind, ITensor *output, Coordinates2D crop_size, - InterpolationPolicy method = InterpolationPolicy::BILINEAR, float extrapolation_value = 0); + void configure(const ITensor *input, + const ITensor *boxes, + const ITensor *box_ind, + ITensor *output, + Coordinates2D crop_size, + InterpolationPolicy method = InterpolationPolicy::BILINEAR, + float extrapolation_value = 0); /** Static function to check if given info will lead to a valid configuration of @ref NESlice * @@ -96,8 +101,13 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *boxes, const ITensorInfo *box_ind, const ITensorInfo *output, - Coordinates2D crop_size, InterpolationPolicy method, float extrapolation_value); + static Status validate(const ITensorInfo *input, + const ITensorInfo *boxes, + const ITensorInfo *box_ind, + const ITensorInfo *output, + Coordinates2D crop_size, + InterpolationPolicy method, + float extrapolation_value); void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h index cdc3a636b0..aabe42f928 100644 --- a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h @@ -24,15 +24,14 @@ #ifndef ARM_COMPUTE_NEDECONVOLUTIONLAYER_H #define ARM_COMPUTE_NEDECONVOLUTIONLAYER_H -#include "arm_compute/runtime/CPP/functions/CPPUpsample.h" -#include "arm_compute/runtime/NEON/functions/NEConvolutionLayer.h" -#include "arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h" -#include "arm_compute/runtime/NEON/functions/NEReverse.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/CPP/functions/CPPUpsample.h" #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" +#include "arm_compute/runtime/NEON/functions/NEConvolutionLayer.h" +#include "arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h" +#include "arm_compute/runtime/NEON/functions/NEReverse.h" #include "arm_compute/runtime/Tensor.h" #include <memory> @@ -117,7 +116,13 @@ public: * the GEMM convolution. * */ - void configure(ITensor *input, const ITensor *weights, const ITensor *bias, ITensor *output, const PadStrideInfo &info, bool enable_fast_math = false, const WeightsInfo &weights_info = WeightsInfo()); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *bias, + ITensor *output, + const PadStrideInfo &info, + bool enable_fast_math = false, + const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. @@ -134,8 +139,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, const ITensorInfo *output, const PadStrideInfo &info, - bool enable_fast_math = false, const WeightsInfo &weights_info = WeightsInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + const ITensorInfo *output, + const PadStrideInfo &info, + bool enable_fast_math = false, + const WeightsInfo &weights_info = WeightsInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEDepthConvertLayer.h b/arm_compute/runtime/NEON/functions/NEDepthConvertLayer.h index eb0724ae12..7bfdfbd13d 100644 --- a/arm_compute/runtime/NEON/functions/NEDepthConvertLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDepthConvertLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEDEPTHCONVERT_H #define ARM_COMPUTE_NEDEPTHCONVERT_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -84,7 +83,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, ConvertPolicy policy, uint32_t shift = 0); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, ConvertPolicy policy, uint32_t shift = 0); // Inherited methods overridden void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEDepthToSpaceLayer.h b/arm_compute/runtime/NEON/functions/NEDepthToSpaceLayer.h index b9bdcd1f11..c7df29a704 100644 --- a/arm_compute/runtime/NEON/functions/NEDepthToSpaceLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDepthToSpaceLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEDEPTHTOSPACELAYER_H #define ARM_COMPUTE_NEDEPTHTOSPACELAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" namespace arm_compute diff --git a/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h index 6f2ec8cddb..6ad5aa7bfa 100644 --- a/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDepthwiseConvolutionLayer.h @@ -28,6 +28,7 @@ #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" #include "arm_compute/runtime/NEON/functions/NEPermute.h" + #include <memory> namespace arm_compute @@ -80,8 +81,14 @@ public: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1). */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); /** Static function to check if given info will lead to a valid configuration of @ref NEDepthwiseConvolutionLayer * @@ -98,8 +105,14 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); // Inherited methods overriden: void run() override; @@ -127,9 +140,11 @@ private: /** Default move constructor */ NEDepthwiseConvolutionLayerOptimizedInternal(NEDepthwiseConvolutionLayerOptimizedInternal &&) = default; /** Prevent instances of this class from being copied (As this class contains pointers) */ - NEDepthwiseConvolutionLayerOptimizedInternal &operator=(const NEDepthwiseConvolutionLayerOptimizedInternal &) = delete; + NEDepthwiseConvolutionLayerOptimizedInternal & + operator=(const NEDepthwiseConvolutionLayerOptimizedInternal &) = delete; /** Default move assignment operator */ - NEDepthwiseConvolutionLayerOptimizedInternal &operator=(NEDepthwiseConvolutionLayerOptimizedInternal &&) = default; + NEDepthwiseConvolutionLayerOptimizedInternal & + operator=(NEDepthwiseConvolutionLayerOptimizedInternal &&) = default; /** Default destructor */ ~NEDepthwiseConvolutionLayerOptimizedInternal() = default; /** Initialize the function's source, destination, kernels and border_size. @@ -144,8 +159,14 @@ private: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1). */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); /** Static function to check if given info will lead to a valid configuration of @ref NEDepthwiseConvolutionLayer3x3 * @@ -161,8 +182,14 @@ private: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); // Inherited methods overriden: void run() override; @@ -207,8 +234,14 @@ private: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1). */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); /** Static function to check if given info will lead to a valid configuration of @ref NEDepthwiseConvolutionLayerGeneric * @@ -225,8 +258,14 @@ private: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - unsigned int depth_multiplier = 1, const ActivationLayerInfo &act_info = ActivationLayerInfo(), const Size2D &dilation = Size2D(1U, 1U)); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + unsigned int depth_multiplier = 1, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + const Size2D &dilation = Size2D(1U, 1U)); // Inherited methods overriden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEDetectionPostProcessLayer.h b/arm_compute/runtime/NEON/functions/NEDetectionPostProcessLayer.h index 2affa8d49e..7a94833d10 100644 --- a/arm_compute/runtime/NEON/functions/NEDetectionPostProcessLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDetectionPostProcessLayer.h @@ -24,13 +24,12 @@ #ifndef ARM_COMPUTE_NE_DETECTION_POSTPROCESS_H #define ARM_COMPUTE_NE_DETECTION_POSTPROCESS_H -#include "arm_compute/runtime/NEON/INESimpleFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/runtime/CPP/functions/CPPDetectionPostProcessLayer.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/NEON/functions/NEDequantizationLayer.h" +#include "arm_compute/runtime/NEON/INESimpleFunction.h" #include "arm_compute/runtime/Tensor.h" #include <map> @@ -78,8 +77,14 @@ public: * * @note Output contains all the detections. Of those, only the ones selected by the valid region are valid. */ - void configure(const ITensor *input_box_encoding, const ITensor *input_score, const ITensor *input_anchors, - ITensor *output_boxes, ITensor *output_classes, ITensor *output_scores, ITensor *num_detection, DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); + void configure(const ITensor *input_box_encoding, + const ITensor *input_score, + const ITensor *input_anchors, + ITensor *output_boxes, + ITensor *output_classes, + ITensor *output_scores, + ITensor *num_detection, + DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEDetectionPostProcessLayer * * @param[in] input_box_encoding The bounding box input tensor info. Data types supported: QASYMM8/QASYMM8_SIGNED/F32. @@ -93,8 +98,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input_box_encoding, const ITensorInfo *input_class_score, const ITensorInfo *input_anchors, - ITensorInfo *output_boxes, ITensorInfo *output_classes, ITensorInfo *output_scores, ITensorInfo *num_detection, + static Status validate(const ITensorInfo *input_box_encoding, + const ITensorInfo *input_class_score, + const ITensorInfo *input_anchors, + ITensorInfo *output_boxes, + ITensorInfo *output_classes, + ITensorInfo *output_scores, + ITensorInfo *num_detection, DetectionPostProcessLayerInfo info = DetectionPostProcessLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h index 8db7e6596b..3ae3b2a15c 100644 --- a/arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDirectConvolutionLayer.h @@ -85,7 +85,12 @@ public: * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. */ - void configure(ITensor *input, const ITensor *weights, const ITensor *bias, ITensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *bias, + ITensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEDirectConvolutionLayer * * @note: DirectConvolution only works in the following configurations: @@ -106,7 +111,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, const ITensorInfo *output, const PadStrideInfo &conv_info, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *bias, + const ITensorInfo *output, + const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: diff --git a/arm_compute/runtime/NEON/functions/NEElementwiseOperations.h b/arm_compute/runtime/NEON/functions/NEElementwiseOperations.h index bfcd221e17..ebf2277d1f 100644 --- a/arm_compute/runtime/NEON/functions/NEElementwiseOperations.h +++ b/arm_compute/runtime/NEON/functions/NEElementwiseOperations.h @@ -73,7 +73,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref cpu::kernels::CpuArithmeticKernel for max * * @param[in] input1 First tensor input info. Data types supported: QASYMM8/QASYMM8_SIGNED/S16/F16/S32/F32. @@ -83,7 +86,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -133,7 +139,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref cpu::kernels::CpuArithmeticKernel for min * * @param[in] input1 First tensor input info. Data types supported: QASYMM8/QASYMM8_SIGNED/S16/F16/S32/F32. @@ -143,7 +152,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -193,7 +205,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref cpu::kernels::CpuArithmeticKernel for squared difference * * @param[in] input1 First tensor input info. Data types supported: QASYMM8/QASYMM8_SIGNED/S16/F16/S32/F32. @@ -203,7 +218,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -249,7 +267,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref cpu::kernels::CpuArithmeticKernel for division * * @param[in] input1 First tensor input info. Data types supported: F16/F32. @@ -259,7 +280,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -306,7 +330,10 @@ public: * @param[out] output Output tensor. Data types supported: Same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref cpu::kernels::CpuArithmeticKernel for power * * @param[in] input1 First tensor input info. Data types supported: F16/F32. @@ -316,7 +343,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; @@ -377,7 +407,8 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ComparisonOperation op); + static Status + validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, ComparisonOperation op); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEFFT1D.h b/arm_compute/runtime/NEON/functions/NEFFT1D.h index 9654b1e604..99c6fd4eb4 100644 --- a/arm_compute/runtime/NEON/functions/NEFFT1D.h +++ b/arm_compute/runtime/NEON/functions/NEFFT1D.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEFFT1D_H #define ARM_COMPUTE_NEFFT1D_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/FunctionDescriptors.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/Tensor.h" diff --git a/arm_compute/runtime/NEON/functions/NEFFT2D.h b/arm_compute/runtime/NEON/functions/NEFFT2D.h index 57f38d1942..cefd3df17a 100644 --- a/arm_compute/runtime/NEON/functions/NEFFT2D.h +++ b/arm_compute/runtime/NEON/functions/NEFFT2D.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEFFT2D_H #define ARM_COMPUTE_NEFFT2D_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/FunctionDescriptors.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/NEON/functions/NEFFT1D.h" #include "arm_compute/runtime/Tensor.h" diff --git a/arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h index c5f4d45b6b..84bfe6b02f 100644 --- a/arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEFFTConvolutionLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEFFTCONVOLUTIONLAYER_H #define ARM_COMPUTE_NEFFTCONVOLUTIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h" #include "arm_compute/runtime/NEON/functions/NEFFT2D.h" @@ -94,8 +93,13 @@ public: * @param[in] act_info (Optional) Activation layer information in case of a fused activation. * @param[in] enable_fast_math (Optional) Enable fast math computation. Unused for CPU backend. */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + void configure(ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); /** Static function to check if given info will lead to a valid configuration of @ref NEFFTConvolutionLayer * * @note: This function only works with any square kernel size and unit strides for both NCHW and NHWC data layout @@ -113,8 +117,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEFill.h b/arm_compute/runtime/NEON/functions/NEFill.h index e923ce33e1..1829c71fef 100644 --- a/arm_compute/runtime/NEON/functions/NEFill.h +++ b/arm_compute/runtime/NEON/functions/NEFill.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_NEFILL_H #define ARM_COMPUTE_NEFILL_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/PixelValue.h" #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NEFillBorder.h b/arm_compute/runtime/NEON/functions/NEFillBorder.h index ab77c28839..44b1d4a62b 100644 --- a/arm_compute/runtime/NEON/functions/NEFillBorder.h +++ b/arm_compute/runtime/NEON/functions/NEFillBorder.h @@ -27,6 +27,7 @@ #include "arm_compute/core/PixelValue.h" #include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -57,7 +58,10 @@ public: * @param[in] border_mode Strategy to use for borders. * @param[in] constant_border_value (Optional) Constant value to use for borders if border_mode is set to CONSTANT. */ - void configure(ITensor *input, unsigned int border_width, BorderMode border_mode, const PixelValue &constant_border_value = PixelValue()); + void configure(ITensor *input, + unsigned int border_width, + BorderMode border_mode, + const PixelValue &constant_border_value = PixelValue()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEFloor.h b/arm_compute/runtime/NEON/functions/NEFloor.h index 4d47b068db..77ac484bab 100644 --- a/arm_compute/runtime/NEON/functions/NEFloor.h +++ b/arm_compute/runtime/NEON/functions/NEFloor.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEFLOOR_H #define ARM_COMPUTE_NEFLOOR_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h b/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h index 05b7ce3735..885f8430cf 100644 --- a/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h +++ b/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h @@ -28,7 +28,6 @@ #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/IWeightsManager.h" - #include "arm_compute/runtime/NEON/functions/NETranspose.h" #include "arm_compute/runtime/Tensor.h" @@ -88,7 +87,8 @@ class NEFullyConnectedLayer : public IFunction { public: /** Constructor */ - NEFullyConnectedLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, IWeightsManager *weights_manager = nullptr); + NEFullyConnectedLayer(std::shared_ptr<IMemoryManager> memory_manager = nullptr, + IWeightsManager *weights_manager = nullptr); /** Prevent instances of this class from being copied (As this class contains pointers) */ NEFullyConnectedLayer(const NEFullyConnectedLayer &) = delete; /** Prevent instances of this class from being moved (As this class contains pointers) */ @@ -126,16 +126,24 @@ public: * @param[in] fc_info (Optional) Fully connected layer additional info * @param[in] weights_info (Optional) Stores neccessary compute information when weights are already reshaped */ - void configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, - FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), const WeightsInfo &weights_info = WeightsInfo()); + void configure(const ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), + const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEFullyConnectedLayer * * Similar to @ref NEFullyConnectedLayer::configure() * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, - FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), const WeightsInfo &weights_info = WeightsInfo()); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + FullyConnectedLayerInfo fc_info = FullyConnectedLayerInfo(), + const WeightsInfo &weights_info = WeightsInfo()); /** Static function that queries whether fixed-format kernel exists for a given problem description * @@ -149,8 +157,13 @@ public: * * @return a status */ - static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, const ITensorInfo *input, const ITensorInfo *weights, - const ITensorInfo *biases, const ITensorInfo *output, const FullyConnectedLayerInfo &fc_info, const WeightsInfo &weights_info); + static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, + const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const FullyConnectedLayerInfo &fc_info, + const WeightsInfo &weights_info); //Inherited methods override void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEFuseBatchNormalization.h b/arm_compute/runtime/NEON/functions/NEFuseBatchNormalization.h index 3dd7f49044..f53b3de7f6 100644 --- a/arm_compute/runtime/NEON/functions/NEFuseBatchNormalization.h +++ b/arm_compute/runtime/NEON/functions/NEFuseBatchNormalization.h @@ -75,9 +75,16 @@ public: * @param[in] epsilon (Optional) Batch normalization layer epsilon parameter. Defaults to 0.001f. * @param[in] fbn_type (Optional) Fused batch normalization type. Defaults to Convolution. */ - void configure(const ITensor *input_weights, const ITensor *bn_mean, const ITensor *bn_var, ITensor *fused_weights, ITensor *fused_bias, - const ITensor *input_bias = nullptr, const ITensor *bn_beta = nullptr, const ITensor *bn_gamma = nullptr, - float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); + void configure(const ITensor *input_weights, + const ITensor *bn_mean, + const ITensor *bn_var, + ITensor *fused_weights, + ITensor *fused_bias, + const ITensor *input_bias = nullptr, + const ITensor *bn_beta = nullptr, + const ITensor *bn_gamma = nullptr, + float epsilon = 0.001f, + FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); /** Static function to check if given info will lead to a valid configuration of @ref NEFuseBatchNormalization * * @param[in] input_weights Input weights tensor info for convolution or depthwise convolution layer. Data type supported: F16/F32. Data layout supported: NCHW, NHWC @@ -95,10 +102,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input_weights, const ITensorInfo *bn_mean, const ITensorInfo *bn_var, - const ITensorInfo *fused_weights, const ITensorInfo *fused_bias, - const ITensorInfo *input_bias = nullptr, const ITensorInfo *bn_beta = nullptr, const ITensorInfo *bn_gamma = nullptr, - float epsilon = 0.001f, FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); + static Status validate(const ITensorInfo *input_weights, + const ITensorInfo *bn_mean, + const ITensorInfo *bn_var, + const ITensorInfo *fused_weights, + const ITensorInfo *fused_bias, + const ITensorInfo *input_bias = nullptr, + const ITensorInfo *bn_beta = nullptr, + const ITensorInfo *bn_gamma = nullptr, + float epsilon = 0.001f, + FuseBatchNormalizationType fbn_type = FuseBatchNormalizationType::CONVOLUTION); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEGEMM.h b/arm_compute/runtime/NEON/functions/NEGEMM.h index c6ff2dfb92..29650a5eca 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMM.h +++ b/arm_compute/runtime/NEON/functions/NEGEMM.h @@ -78,14 +78,26 @@ public: * @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and * if the reshape of matrix B should happen only for the first run */ - void configure(const ITensor *a, const ITensor *b, const ITensor *c, ITensor *d, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + void configure(const ITensor *a, + const ITensor *b, + const ITensor *c, + ITensor *d, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEGEMM. * * Similar to @ref NEGEMM::configure() * * @return a status */ - static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + static Status validate(const ITensorInfo *a, + const ITensorInfo *b, + const ITensorInfo *c, + const ITensorInfo *output, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); /** Static function that queries whether there exists fixed-format kernel and if it exists it will return in the first argument in what format * weights are expected to be reshaped as defined by WeightFormat class. Apart from the first argument the rest of the arguments are the same @@ -93,8 +105,14 @@ public: * * @return a status */ - static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, - float alpha, float beta, const GEMMInfo &gemm_info = GEMMInfo()); + static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, + const ITensorInfo *a, + const ITensorInfo *b, + const ITensorInfo *c, + const ITensorInfo *output, + float alpha, + float beta, + const GEMMInfo &gemm_info = GEMMInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEGEMMConv2d.h b/arm_compute/runtime/NEON/functions/NEGEMMConv2d.h index 53ceb6d978..d1c5a1c9b3 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMConv2d.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMConv2d.h @@ -86,7 +86,8 @@ public: * Data types supported: Same as @p input. * @param[in] info Convolution layer descriptor */ - void configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const Conv2dInfo &info); + void + configure(ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const Conv2dInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMConv2d * * @param[in] input Source tensor info. 3 lower dimensions represent a single input [width, height, IFM], @@ -102,7 +103,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const Conv2dInfo &info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const Conv2dInfo &info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h index 72309e464e..3e84c3e2cf 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMConvolutionLayer.h @@ -24,8 +24,6 @@ #ifndef ARM_COMPUTE_NEGEMMCONVOLUTIONLAYER_H #define ARM_COMPUTE_NEGEMMCONVOLUTIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" #include "arm_compute/runtime/IFunction.h" @@ -49,7 +47,8 @@ class NEGEMMConvolutionLayer : public IFunction { public: /** Constructor */ - NEGEMMConvolutionLayer(const std::shared_ptr<IMemoryManager> &memory_manager = nullptr, IWeightsManager *weights_manager = nullptr); + NEGEMMConvolutionLayer(const std::shared_ptr<IMemoryManager> &memory_manager = nullptr, + IWeightsManager *weights_manager = nullptr); /** Prevent instances of this class from being copied (As this class contains pointers) */ NEGEMMConvolutionLayer(const NEGEMMConvolutionLayer &) = delete; /** Prevent instances of this class from being moved (As this class contains non movable objects) */ @@ -95,8 +94,16 @@ public: * available which may introduce a drop of accuracy as well. Default is false * @param[in] num_groups (Optional) Number of groups when performing a grouped convolution. num_groups != 1 is not supported */ - void configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const WeightsInfo &weights_info = WeightsInfo(), - const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false, unsigned int num_groups = 1); + void configure(const ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMConvolutionLayer * * @param[in] input Source tensor info. 3 lower dimensions represent a single input [width, height, IFM], @@ -119,9 +126,16 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), - bool enable_fast_math = false, unsigned int num_groups = 1); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false, + unsigned int num_groups = 1); /** Static function to check if there is an optimized version of * GEMM available for the input parameters. @@ -178,10 +192,16 @@ public: * * @return a Status */ - static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, const ITensorInfo *src, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *dst, - const PadStrideInfo &conv_info, - const WeightsInfo &weights_info = WeightsInfo(), const Size2D &dilation = Size2D(1U, 1U), const ActivationLayerInfo &act_info = ActivationLayerInfo(), - bool enable_fast_math = false); + static Status has_opt_impl(arm_compute::WeightFormat &expected_weight_format, + const ITensorInfo *src, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *dst, + const PadStrideInfo &conv_info, + const WeightsInfo &weights_info = WeightsInfo(), + const Size2D &dilation = Size2D(1U, 1U), + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; void prepare() override; diff --git a/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h b/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h index addb13cdfa..824c4443ad 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMLowpMatrixMultiplyCore.h @@ -47,7 +47,8 @@ class NEGEMMLowpMatrixMultiplyCore : public IFunction { public: /** Constructor */ - NEGEMMLowpMatrixMultiplyCore(std::shared_ptr<IMemoryManager> memory_manager = nullptr, IWeightsManager *weights_manager = nullptr); + NEGEMMLowpMatrixMultiplyCore(std::shared_ptr<IMemoryManager> memory_manager = nullptr, + IWeightsManager *weights_manager = nullptr); /** Prevent instances of this class from being copied (As this class contains pointers) */ NEGEMMLowpMatrixMultiplyCore(const NEGEMMLowpMatrixMultiplyCore &) = delete; /** Default move constructor */ @@ -96,14 +97,19 @@ public: * @param[in] gemm_info (Optional) Specifies if the matrix A and/or matrix B have been reshaped and * if the reshape of matrix B should be executed only for the first run */ - void configure(const ITensor *a, const ITensor *b, const ITensor *c, ITensor *output, const GEMMInfo &gemm_info = GEMMInfo()); + void configure( + const ITensor *a, const ITensor *b, const ITensor *c, ITensor *output, const GEMMInfo &gemm_info = GEMMInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEGEMMLowpMatrixMultiplyCore * * Similar to @ref NEGEMMLowpMatrixMultiplyCore::configure() * * @return a status */ - static Status validate(const ITensorInfo *a, const ITensorInfo *b, const ITensorInfo *c, const ITensorInfo *output, const GEMMInfo &gemm_info = GEMMInfo()); + static Status validate(const ITensorInfo *a, + const ITensorInfo *b, + const ITensorInfo *c, + const ITensorInfo *output, + const GEMMInfo &gemm_info = GEMMInfo()); // Inherited methods overridden void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h b/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h index 232344e5c2..0d932bb4af 100644 --- a/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h +++ b/arm_compute/runtime/NEON/functions/NEGEMMLowpOutputStage.h @@ -89,7 +89,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *bias, const ITensorInfo *output, const GEMMLowpOutputStageInfo &info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *bias, + const ITensorInfo *output, + const GEMMLowpOutputStageInfo &info); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEGenerateProposalsLayer.h b/arm_compute/runtime/NEON/functions/NEGenerateProposalsLayer.h index 3b683382ec..0f294fde22 100644 --- a/arm_compute/runtime/NEON/functions/NEGenerateProposalsLayer.h +++ b/arm_compute/runtime/NEON/functions/NEGenerateProposalsLayer.h @@ -95,7 +95,12 @@ public: * @note Only single image prediction is supported. Height and Width (and scale) of the image will be contained in the @ref GenerateProposalsInfo struct. * @note Proposals contains all the proposals. Of those, only the first num_valid_proposals are valid. */ - void configure(const ITensor *scores, const ITensor *deltas, const ITensor *anchors, ITensor *proposals, ITensor *scores_out, ITensor *num_valid_proposals, + void configure(const ITensor *scores, + const ITensor *deltas, + const ITensor *anchors, + ITensor *proposals, + ITensor *scores_out, + ITensor *num_valid_proposals, const GenerateProposalsInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref NEGenerateProposalsLayer @@ -112,7 +117,11 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *scores, const ITensorInfo *deltas, const ITensorInfo *anchors, const ITensorInfo *proposals, const ITensorInfo *scores_out, + static Status validate(const ITensorInfo *scores, + const ITensorInfo *deltas, + const ITensorInfo *anchors, + const ITensorInfo *proposals, + const ITensorInfo *scores_out, const ITensorInfo *num_valid_proposals, const GenerateProposalsInfo &info); diff --git a/arm_compute/runtime/NEON/functions/NEInstanceNormalizationLayer.h b/arm_compute/runtime/NEON/functions/NEInstanceNormalizationLayer.h index bb0697072b..0bc57be09e 100644 --- a/arm_compute/runtime/NEON/functions/NEInstanceNormalizationLayer.h +++ b/arm_compute/runtime/NEON/functions/NEInstanceNormalizationLayer.h @@ -89,7 +89,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, float gamma = 1.0f, float beta = 0.0f, float epsilon = 1e-12f); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + float gamma = 1.0f, + float beta = 0.0f, + float epsilon = 1e-12f); // Inherited methods overridden: void run() override; @@ -103,5 +107,5 @@ private: Tensor _permuted_input; Tensor _permuted_output; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEINSTANCENORMALIZATIONLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEL2NormalizeLayer.h b/arm_compute/runtime/NEON/functions/NEL2NormalizeLayer.h index 7f1a5e785e..8502cee5d2 100644 --- a/arm_compute/runtime/NEON/functions/NEL2NormalizeLayer.h +++ b/arm_compute/runtime/NEON/functions/NEL2NormalizeLayer.h @@ -97,5 +97,5 @@ private: std::unique_ptr<NEL2NormalizeLayerKernel> _normalize_kernel; Tensor _sumsq; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEL2NORMALIZELAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NELSTMLayer.h b/arm_compute/runtime/NEON/functions/NELSTMLayer.h index 4272215486..629c5d10a0 100644 --- a/arm_compute/runtime/NEON/functions/NELSTMLayer.h +++ b/arm_compute/runtime/NEON/functions/NELSTMLayer.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_NELSTMLAYER_H #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/common/LSTMParams.h" #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h" @@ -35,7 +36,6 @@ #include "arm_compute/runtime/NEON/functions/NEMeanStdDevNormalizationLayer.h" #include "arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h" #include "arm_compute/runtime/NEON/functions/NETranspose.h" -#include "arm_compute/runtime/common/LSTMParams.h" namespace arm_compute { @@ -104,13 +104,26 @@ public: * @param[in] projection_threshold The clipping threshold for the output from the projection layer, such that values are bound within [-proj_clip, proj_clip]. * If set to 0.0 then clipping is disabled. */ - void configure(const ITensor *input, - const ITensor *input_to_forget_weights, const ITensor *input_to_cell_weights, const ITensor *input_to_output_weights, - const ITensor *recurrent_to_forget_weights, const ITensor *recurrent_to_cell_weights, const ITensor *recurrent_to_output_weights, - const ITensor *forget_gate_bias, const ITensor *cell_bias, const ITensor *output_gate_bias, - const ITensor *output_state_in, const ITensor *cell_state_in, - ITensor *scratch_buffer, ITensor *output_state_out, ITensor *cell_state_out, ITensor *output, - const LSTMParams<ITensor> &lstm_params, const ActivationLayerInfo &activation_info, float cell_threshold = 0.f, float projection_threshold = 0.f); + void configure(const ITensor *input, + const ITensor *input_to_forget_weights, + const ITensor *input_to_cell_weights, + const ITensor *input_to_output_weights, + const ITensor *recurrent_to_forget_weights, + const ITensor *recurrent_to_cell_weights, + const ITensor *recurrent_to_output_weights, + const ITensor *forget_gate_bias, + const ITensor *cell_bias, + const ITensor *output_gate_bias, + const ITensor *output_state_in, + const ITensor *cell_state_in, + ITensor *scratch_buffer, + ITensor *output_state_out, + ITensor *cell_state_out, + ITensor *output, + const LSTMParams<ITensor> &lstm_params, + const ActivationLayerInfo &activation_info, + float cell_threshold = 0.f, + float projection_threshold = 0.f); /** Static function to check if given info will lead to a valid configuration of @ref NELSTMLayer * @@ -151,13 +164,26 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *output_state_in, const ITensorInfo *cell_state_in, - const ITensorInfo *scratch_buffer, const ITensorInfo *output_state_out, const ITensorInfo *cell_state_out, const ITensorInfo *output, - const LSTMParams<ITensorInfo> &lstm_params, const ActivationLayerInfo &activation_info, float cell_threshold = 0.f, float projection_threshold = 0.f); + static Status validate(const ITensorInfo *input, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_in, + const ITensorInfo *scratch_buffer, + const ITensorInfo *output_state_out, + const ITensorInfo *cell_state_out, + const ITensorInfo *output, + const LSTMParams<ITensorInfo> &lstm_params, + const ActivationLayerInfo &activation_info, + float cell_threshold = 0.f, + float projection_threshold = 0.f); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h b/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h index bcb89d997d..ae951669b3 100644 --- a/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h +++ b/arm_compute/runtime/NEON/functions/NELSTMLayerQuantized.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_NELSTMLAYERQUANTIZED_H #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/common/LSTMParams.h" #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h" #include "arm_compute/runtime/NEON/functions/NEConcatenateLayer.h" @@ -38,8 +39,6 @@ #include "arm_compute/runtime/NEON/functions/NESlice.h" #include "arm_compute/runtime/NEON/functions/NETranspose.h" -#include "arm_compute/runtime/common/LSTMParams.h" - namespace arm_compute { // Forward declarations @@ -104,11 +103,22 @@ public: * @param[out] output_state_out Destination tensor. Output is a 2D tensor with dimensions [output_size, batch_size].Data types supported: Same as @p input. */ void configure(const ITensor *input, - const ITensor *input_to_input_weights, const ITensor *input_to_forget_weights, const ITensor *input_to_cell_weights, const ITensor *input_to_output_weights, - const ITensor *recurrent_to_input_weights, const ITensor *recurrent_to_forget_weights, const ITensor *recurrent_to_cell_weights, const ITensor *recurrent_to_output_weights, - const ITensor *input_gate_bias, const ITensor *forget_gate_bias, const ITensor *cell_bias, const ITensor *output_gate_bias, - ITensor *cell_state_in, const ITensor *output_state_in, - ITensor *cell_state_out, ITensor *output_state_out); + const ITensor *input_to_input_weights, + const ITensor *input_to_forget_weights, + const ITensor *input_to_cell_weights, + const ITensor *input_to_output_weights, + const ITensor *recurrent_to_input_weights, + const ITensor *recurrent_to_forget_weights, + const ITensor *recurrent_to_cell_weights, + const ITensor *recurrent_to_output_weights, + const ITensor *input_gate_bias, + const ITensor *forget_gate_bias, + const ITensor *cell_bias, + const ITensor *output_gate_bias, + ITensor *cell_state_in, + const ITensor *output_state_in, + ITensor *cell_state_out, + ITensor *output_state_out); /** Static function to check if given info will lead to a valid configuration of @ref NELSTMLayer * @@ -133,11 +143,22 @@ public: * @return a status */ static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_input_weights, const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_input_weights, const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *input_gate_bias, const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, - const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out); + const ITensorInfo *input_to_input_weights, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_input_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *input_gate_bias, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, + const ITensorInfo *output_state_out); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEMatMul.h b/arm_compute/runtime/NEON/functions/NEMatMul.h index e961f860c1..414fc2f3fd 100644 --- a/arm_compute/runtime/NEON/functions/NEMatMul.h +++ b/arm_compute/runtime/NEON/functions/NEMatMul.h @@ -27,6 +27,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -48,7 +49,7 @@ public: }; private: - bool _fast_math{ false }; + bool _fast_math{false}; }; // Forward declarations @@ -96,7 +97,12 @@ public: * @param[in] settings Contains flags for function level settings i.e fast math * @param[in] act_info (Optional) Contains activation function and lower and upper bound values for bounded activation functions. */ - void configure(ITensor *lhs, ITensor *rhs, ITensor *dst, const MatMulInfo &info, const CpuMatMulSettings &settings, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *lhs, + ITensor *rhs, + ITensor *dst, + const MatMulInfo &info, + const CpuMatMulSettings &settings, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEMatMul * * @param[in] lhs Left-hand side tensor info. Data types supported: F16/F32/QASYMM8_SIGNED/QASYMM8. @@ -108,7 +114,11 @@ public: * * @return Status */ - static Status validate(const ITensorInfo *lhs, const ITensorInfo *rhs, const ITensorInfo *dst, const MatMulInfo &info, const CpuMatMulSettings &settings, + static Status validate(const ITensorInfo *lhs, + const ITensorInfo *rhs, + const ITensorInfo *dst, + const MatMulInfo &info, + const CpuMatMulSettings &settings, const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden @@ -118,5 +128,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEMATMUL */ diff --git a/arm_compute/runtime/NEON/functions/NEMaxUnpoolingLayer.h b/arm_compute/runtime/NEON/functions/NEMaxUnpoolingLayer.h index 2f77540e1e..e00fc4544f 100644 --- a/arm_compute/runtime/NEON/functions/NEMaxUnpoolingLayer.h +++ b/arm_compute/runtime/NEON/functions/NEMaxUnpoolingLayer.h @@ -26,6 +26,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -86,7 +87,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *indices, const ITensorInfo *output, const PoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *indices, + const ITensorInfo *output, + const PoolingLayerInfo &pool_info); // Inherited methods overridden: void run() override; @@ -96,5 +100,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEMAXUNPOOLINGLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NENormalizationLayer.h b/arm_compute/runtime/NEON/functions/NENormalizationLayer.h index fbe000445c..27e3fa674e 100644 --- a/arm_compute/runtime/NEON/functions/NENormalizationLayer.h +++ b/arm_compute/runtime/NEON/functions/NENormalizationLayer.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NENORMALIZATIONLAYER_H #define ARM_COMPUTE_NENORMALIZATIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/IMemoryManager.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h" @@ -88,16 +87,17 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const NormalizationLayerInfo &norm_info); // Inherited methods overridden: void run() override; private: - MemoryGroup _memory_group; /**< Function memory group */ - std::unique_ptr<NENormalizationLayerKernel> _norm_kernel; /**< Normalization layer kernel */ - NEPixelWiseMultiplication _multiply_f; /**< Pixel multiplication function */ - Tensor _input_squared; /**< The intermediate buffer which stores results of squaring input */ + MemoryGroup _memory_group; /**< Function memory group */ + std::unique_ptr<NENormalizationLayerKernel> _norm_kernel; /**< Normalization layer kernel */ + NEPixelWiseMultiplication _multiply_f; /**< Pixel multiplication function */ + Tensor _input_squared; /**< The intermediate buffer which stores results of squaring input */ }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NENORMALIZATIONLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEPadLayer.h b/arm_compute/runtime/NEON/functions/NEPadLayer.h index 4aa6725496..494b1c0641 100644 --- a/arm_compute/runtime/NEON/functions/NEPadLayer.h +++ b/arm_compute/runtime/NEON/functions/NEPadLayer.h @@ -24,14 +24,14 @@ #ifndef ARM_COMPUTE_NEPADLAYER_H #define ARM_COMPUTE_NEPADLAYER_H +#include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/NEON/functions/NEConcatenateLayer.h" #include "arm_compute/runtime/NEON/functions/NECopy.h" #include "arm_compute/runtime/NEON/functions/NEStridedSlice.h" #include "arm_compute/runtime/SubTensor.h" - -#include "arm_compute/core/Types.h" #include "arm_compute/runtime/Tensor.h" + #include <memory> namespace arm_compute @@ -82,7 +82,11 @@ public: * @param[in] mode (Optional) Controls whether the padding should be filled with @p constant_value using CONSTANT, * or reflect the input, either including the border values (SYMMETRIC) or not (REFLECT). */ - void configure(ITensor *input, ITensor *output, const PaddingList &padding, const PixelValue constant_value = PixelValue(), const PaddingMode mode = PaddingMode::CONSTANT); + void configure(ITensor *input, + ITensor *output, + const PaddingList &padding, + const PixelValue constant_value = PixelValue(), + const PaddingMode mode = PaddingMode::CONSTANT); /** Static function to check if given info will lead to a valid configuration of @ref NEPadLayer. * * @param[in] input Source tensor info. Data types supported: All. @@ -95,7 +99,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PaddingList &padding, const PixelValue constant_value = PixelValue(), const PaddingMode mode = PaddingMode::CONSTANT); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const PaddingList &padding, + const PixelValue constant_value = PixelValue(), + const PaddingMode mode = PaddingMode::CONSTANT); // Inherited methods overridden: void run() override; @@ -109,7 +117,10 @@ private: * specifies the front and the end padding in the i-th dimension. * @param[in] constant_value Constant value to be used for the padding */ - void configure_constant_mode(ITensor *input, ITensor *output, const PaddingList &padding, const PixelValue constant_value); + void configure_constant_mode(ITensor *input, + ITensor *output, + const PaddingList &padding, + const PixelValue constant_value); /** Configure functions for when reflect or symmetric padding is used. * * @param[in] input Source tensor. Data types supported: All. diff --git a/arm_compute/runtime/NEON/functions/NEPermute.h b/arm_compute/runtime/NEON/functions/NEPermute.h index c863fde0ac..2cef64764d 100644 --- a/arm_compute/runtime/NEON/functions/NEPermute.h +++ b/arm_compute/runtime/NEON/functions/NEPermute.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEPERMUTE_H #define ARM_COMPUTE_NEPERMUTE_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h b/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h index 634e8e0c39..3d81bf6087 100644 --- a/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h +++ b/arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h @@ -95,7 +95,12 @@ public: * @param[in] rounding_policy Rounding policy. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(const ITensor *input1, const ITensor *input2, ITensor *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy, + void configure(const ITensor *input1, + const ITensor *input2, + ITensor *output, + float scale, + ConvertPolicy overflow_policy, + RoundingPolicy rounding_policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEPixelWiseMultiplication * @@ -122,7 +127,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, float scale, ConvertPolicy overflow_policy, RoundingPolicy rounding_policy, + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + float scale, + ConvertPolicy overflow_policy, + RoundingPolicy rounding_policy, const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: @@ -158,7 +168,10 @@ public: * @param[out] output The output tensor. Data types supported: same as @p input1. Number of channels: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - void configure(ITensor *input1, ITensor *input2, ITensor *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + void configure(ITensor *input1, + ITensor *input2, + ITensor *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEComplexPixelWiseMultiplication * * @param[in] input1 An input tensor info. Data types supported: F32. Number of channels supported: 2 (complex tensor). @@ -166,7 +179,10 @@ public: * @param[in] output The output tensor info. Data types supported: same as @p input1. Number of channels supported: same as @p input1. * @param[in] act_info (Optional) Activation layer information in case of a fused activation. Currently not supported. */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const ActivationLayerInfo &act_info = ActivationLayerInfo()); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const ActivationLayerInfo &act_info = ActivationLayerInfo()); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEPooling3dLayer.h b/arm_compute/runtime/NEON/functions/NEPooling3dLayer.h index 4c5eb58e05..09251f2a5f 100644 --- a/arm_compute/runtime/NEON/functions/NEPooling3dLayer.h +++ b/arm_compute/runtime/NEON/functions/NEPooling3dLayer.h @@ -92,5 +92,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEPOOLING3DLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEPoolingLayer.h b/arm_compute/runtime/NEON/functions/NEPoolingLayer.h index 9147ad9687..768ad0d818 100644 --- a/arm_compute/runtime/NEON/functions/NEPoolingLayer.h +++ b/arm_compute/runtime/NEON/functions/NEPoolingLayer.h @@ -91,7 +91,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const PoolingLayerInfo &pool_info, const ITensorInfo *indices = nullptr); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const PoolingLayerInfo &pool_info, + const ITensorInfo *indices = nullptr); // Inherited methods overridden: void run() override; @@ -100,5 +103,5 @@ private: struct Impl; std::unique_ptr<Impl> _impl; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEPOOLINGLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEPriorBoxLayer.h b/arm_compute/runtime/NEON/functions/NEPriorBoxLayer.h index 38e0c9f3ad..858e3299af 100644 --- a/arm_compute/runtime/NEON/functions/NEPriorBoxLayer.h +++ b/arm_compute/runtime/NEON/functions/NEPriorBoxLayer.h @@ -62,7 +62,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input1, const ITensorInfo *input2, const ITensorInfo *output, const PriorBoxLayerInfo &info); + static Status validate(const ITensorInfo *input1, + const ITensorInfo *input2, + const ITensorInfo *output, + const PriorBoxLayerInfo &info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_NEPRIORBOXLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h b/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h index 185d821ec0..009a4e0911 100644 --- a/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h +++ b/arm_compute/runtime/NEON/functions/NEQLSTMLayer.h @@ -25,6 +25,7 @@ #define ARM_COMPUTE_NEQLSTMLAYER_H #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/common/LSTMParams.h" #include "arm_compute/runtime/NEON/functions/NEActivationLayer.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticAddition.h" #include "arm_compute/runtime/NEON/functions/NEArithmeticSubtraction.h" @@ -35,7 +36,6 @@ #include "arm_compute/runtime/NEON/functions/NEPixelWiseMultiplication.h" #include "arm_compute/runtime/NEON/functions/NEQuantizationLayer.h" #include "arm_compute/runtime/NEON/functions/NETranspose.h" -#include "arm_compute/runtime/common/LSTMParams.h" #include <memory> @@ -130,12 +130,21 @@ public: * projection_threshold (Optional) The clipping threshold for the output from the projection layer, such that values are bound within * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. */ - void configure(const ITensor *input, - const ITensor *input_to_forget_weights, const ITensor *input_to_cell_weights, const ITensor *input_to_output_weights, - const ITensor *recurrent_to_forget_weights, const ITensor *recurrent_to_cell_weights, const ITensor *recurrent_to_output_weights, - const ITensor *forget_gate_bias, const ITensor *cell_bias, const ITensor *output_gate_bias, - const ITensor *cell_state_in, ITensor *output_state_in, - ITensor *cell_state_out, ITensor *output_state_out, ITensor *output, + void configure(const ITensor *input, + const ITensor *input_to_forget_weights, + const ITensor *input_to_cell_weights, + const ITensor *input_to_output_weights, + const ITensor *recurrent_to_forget_weights, + const ITensor *recurrent_to_cell_weights, + const ITensor *recurrent_to_output_weights, + const ITensor *forget_gate_bias, + const ITensor *cell_bias, + const ITensor *output_gate_bias, + const ITensor *cell_state_in, + ITensor *output_state_in, + ITensor *cell_state_out, + ITensor *output_state_out, + ITensor *output, const LSTMParams<ITensor> &lstm_params); /** Static function to check if given info will lead to a valid configuration of @ref NEQLSTMLayer @@ -180,12 +189,21 @@ public: * [-proj_clip, proj_clip]. If set to 0.0 then clipping is disabled. * @return a status */ - static Status validate(const ITensorInfo *input, - const ITensorInfo *input_to_forget_weights, const ITensorInfo *input_to_cell_weights, const ITensorInfo *input_to_output_weights, - const ITensorInfo *recurrent_to_forget_weights, const ITensorInfo *recurrent_to_cell_weights, const ITensorInfo *recurrent_to_output_weights, - const ITensorInfo *forget_gate_bias, const ITensorInfo *cell_bias, const ITensorInfo *output_gate_bias, - const ITensorInfo *cell_state_in, const ITensorInfo *output_state_in, - const ITensorInfo *cell_state_out, const ITensorInfo *output_state_out, const ITensorInfo *output, + static Status validate(const ITensorInfo *input, + const ITensorInfo *input_to_forget_weights, + const ITensorInfo *input_to_cell_weights, + const ITensorInfo *input_to_output_weights, + const ITensorInfo *recurrent_to_forget_weights, + const ITensorInfo *recurrent_to_cell_weights, + const ITensorInfo *recurrent_to_output_weights, + const ITensorInfo *forget_gate_bias, + const ITensorInfo *cell_bias, + const ITensorInfo *output_gate_bias, + const ITensorInfo *cell_state_in, + const ITensorInfo *output_state_in, + const ITensorInfo *cell_state_out, + const ITensorInfo *output_state_out, + const ITensorInfo *output, const LSTMParams<ITensorInfo> &lstm_params); // Inherited methods overridden: @@ -218,10 +236,17 @@ private: * @param[in] mm_res_info Tensor info to be used to initialize output stage result tensor. * */ - void configure_mm(NEGEMMLowpMatrixMultiplyCore &mm, NEGEMMLowpOutputStage &outstage, GEMMLowpOutputStageInfo &gemmlowp_info, - const ITensor *mm_input, const ITensor *mm_weights, const ITensor *bias, Tensor *mm_res, - Tensor *outstage_res, float gemmlowp_scale, - const TensorInfo &mm_res_info, const TensorInfo &outstage_tensor_info); + void configure_mm(NEGEMMLowpMatrixMultiplyCore &mm, + NEGEMMLowpOutputStage &outstage, + GEMMLowpOutputStageInfo &gemmlowp_info, + const ITensor *mm_input, + const ITensor *mm_weights, + const ITensor *bias, + Tensor *mm_res, + Tensor *outstage_res, + float gemmlowp_scale, + const TensorInfo &mm_res_info, + const TensorInfo &outstage_tensor_info); MemoryGroup _memory_group; @@ -230,8 +255,8 @@ private: { static constexpr uint32_t max_dimension_supported = 2; - ITensor *_src{ nullptr }; - ITensor *_dst{ nullptr }; + ITensor *_src{nullptr}; + ITensor *_dst{nullptr}; size_t _row_size{}; Window _window{}; @@ -335,19 +360,16 @@ private: NECopy _copy_output; // Tensor pointers - const ITensor *_input_to_input_weights - { - nullptr - }; - const ITensor *_recurrent_to_input_weights{ nullptr }; - const ITensor *_projection_bias{ nullptr }; - const ITensor *_input_to_forget_weights{ nullptr }; - const ITensor *_input_to_cell_weights{ nullptr }; - const ITensor *_input_to_output_weights{ nullptr }; - const ITensor *_recurrent_to_forget_weights{ nullptr }; - const ITensor *_recurrent_to_cell_weights{ nullptr }; - const ITensor *_recurrent_to_output_weights{ nullptr }; - const ITensor *_projection_weights{ nullptr }; + const ITensor *_input_to_input_weights{nullptr}; + const ITensor *_recurrent_to_input_weights{nullptr}; + const ITensor *_projection_bias{nullptr}; + const ITensor *_input_to_forget_weights{nullptr}; + const ITensor *_input_to_cell_weights{nullptr}; + const ITensor *_input_to_output_weights{nullptr}; + const ITensor *_recurrent_to_forget_weights{nullptr}; + const ITensor *_recurrent_to_cell_weights{nullptr}; + const ITensor *_recurrent_to_output_weights{nullptr}; + const ITensor *_projection_weights{nullptr}; std::array<const ITensor *, _layer_norm_count> _layer_norm_weights{}; std::array<const ITensor *, _layer_norm_count> _layer_norm_bias{}; @@ -382,66 +404,66 @@ private: return _layer_norms[getGateIndex(g)]; } - void configure_layer_norm(LayerNormGate g, const ITensor *in); + void configure_layer_norm(LayerNormGate g, const ITensor *in); static Status validate_layer_norm(const ITensorInfo &in, const ITensorInfo &weight, const ITensorInfo &bias); // Temporary tensors - Tensor _input_to_forget_weights_f32{ nullptr }; - Tensor _input_to_forget_weights_symm8{ nullptr }; + Tensor _input_to_forget_weights_f32{nullptr}; + Tensor _input_to_forget_weights_symm8{nullptr}; - Tensor _input_to_forget_weights_transposed{ nullptr }; - Tensor _input_to_cell_weights_transposed{ nullptr }; - Tensor _input_to_output_weights_transposed{ nullptr }; - Tensor _input_to_input_weights_transposed{ nullptr }; - Tensor _recurrent_to_forget_weights_transposed{ nullptr }; - Tensor _recurrent_to_cell_weights_transposed{ nullptr }; - Tensor _recurrent_to_output_weights_transposed{ nullptr }; - Tensor _recurrent_to_input_weights_transposed{ nullptr }; - Tensor _projection_weights_transposed{ nullptr }; - Tensor _input_to_input_eff_bias{ nullptr }; - Tensor _recurrent_to_input_eff_bias{ nullptr }; - Tensor _input_to_forget_eff_bias{ nullptr }; - Tensor _recurrent_to_forget_eff_bias{ nullptr }; - Tensor _input_to_cell_eff_bias{ nullptr }; - Tensor _recurrent_to_cell_eff_bias{ nullptr }; - Tensor _input_to_output_eff_bias{ nullptr }; - Tensor _recurrent_to_output_eff_bias{ nullptr }; - Tensor _projection_reduction_res{ nullptr }; - Tensor _projection_eff_bias{ nullptr }; - Tensor _mm_input_to_forget_res{ nullptr }; - Tensor _mm_recurrent_to_forget_res{ nullptr }; - Tensor _mul_cell_to_forget_res{ nullptr }; - Tensor _input_to_forget_outstage_res{ nullptr }; - Tensor _cell_to_forget_outstage_res{ nullptr }; - Tensor _recurrent_to_forget_outstage_res{ nullptr }; - Tensor _forget_gate{ nullptr }; - Tensor _mm_input_to_cell_res{ nullptr }; - Tensor _input_to_cell_outstage_res{ nullptr }; - Tensor _mm_recurrent_to_cell_res{ nullptr }; - Tensor _recurrent_to_cell_outstage_res{ nullptr }; - Tensor _cell_gate{ nullptr }; - Tensor _mul_input_cell_res{ nullptr }; - Tensor _mm_input_to_input_res{ nullptr }; - Tensor _input_to_input_outstage_res{ nullptr }; - Tensor _mm_recurrent_to_input_res{ nullptr }; - Tensor _mul_cell_to_input_res{ nullptr }; - Tensor _cell_to_input_outstage_res{ nullptr }; - Tensor _recurrent_to_input_outstage_res{ nullptr }; - Tensor _input_gate{ nullptr }; - Tensor _mm_input_to_output_res{ nullptr }; - Tensor _input_to_output_outstage_res{ nullptr }; - Tensor _mm_recurrent_to_output_res{ nullptr }; - Tensor _mul_cell_to_output_res{ nullptr }; - Tensor _cell_to_output_outstage_res{ nullptr }; - Tensor _recurrent_to_output_outstage_res{ nullptr }; - Tensor _output_gate{ nullptr }; - Tensor _hidden_mul_res{ nullptr }; - Tensor _hidden_gate{ nullptr }; - Tensor _mm_projection_res{ nullptr }; - Tensor _projection_outstage_res{ nullptr }; - Tensor _projection_out_res{ nullptr }; - Tensor _projection_accumulate_res{ nullptr }; - Tensor _ones{ nullptr }; + Tensor _input_to_forget_weights_transposed{nullptr}; + Tensor _input_to_cell_weights_transposed{nullptr}; + Tensor _input_to_output_weights_transposed{nullptr}; + Tensor _input_to_input_weights_transposed{nullptr}; + Tensor _recurrent_to_forget_weights_transposed{nullptr}; + Tensor _recurrent_to_cell_weights_transposed{nullptr}; + Tensor _recurrent_to_output_weights_transposed{nullptr}; + Tensor _recurrent_to_input_weights_transposed{nullptr}; + Tensor _projection_weights_transposed{nullptr}; + Tensor _input_to_input_eff_bias{nullptr}; + Tensor _recurrent_to_input_eff_bias{nullptr}; + Tensor _input_to_forget_eff_bias{nullptr}; + Tensor _recurrent_to_forget_eff_bias{nullptr}; + Tensor _input_to_cell_eff_bias{nullptr}; + Tensor _recurrent_to_cell_eff_bias{nullptr}; + Tensor _input_to_output_eff_bias{nullptr}; + Tensor _recurrent_to_output_eff_bias{nullptr}; + Tensor _projection_reduction_res{nullptr}; + Tensor _projection_eff_bias{nullptr}; + Tensor _mm_input_to_forget_res{nullptr}; + Tensor _mm_recurrent_to_forget_res{nullptr}; + Tensor _mul_cell_to_forget_res{nullptr}; + Tensor _input_to_forget_outstage_res{nullptr}; + Tensor _cell_to_forget_outstage_res{nullptr}; + Tensor _recurrent_to_forget_outstage_res{nullptr}; + Tensor _forget_gate{nullptr}; + Tensor _mm_input_to_cell_res{nullptr}; + Tensor _input_to_cell_outstage_res{nullptr}; + Tensor _mm_recurrent_to_cell_res{nullptr}; + Tensor _recurrent_to_cell_outstage_res{nullptr}; + Tensor _cell_gate{nullptr}; + Tensor _mul_input_cell_res{nullptr}; + Tensor _mm_input_to_input_res{nullptr}; + Tensor _input_to_input_outstage_res{nullptr}; + Tensor _mm_recurrent_to_input_res{nullptr}; + Tensor _mul_cell_to_input_res{nullptr}; + Tensor _cell_to_input_outstage_res{nullptr}; + Tensor _recurrent_to_input_outstage_res{nullptr}; + Tensor _input_gate{nullptr}; + Tensor _mm_input_to_output_res{nullptr}; + Tensor _input_to_output_outstage_res{nullptr}; + Tensor _mm_recurrent_to_output_res{nullptr}; + Tensor _mul_cell_to_output_res{nullptr}; + Tensor _cell_to_output_outstage_res{nullptr}; + Tensor _recurrent_to_output_outstage_res{nullptr}; + Tensor _output_gate{nullptr}; + Tensor _hidden_mul_res{nullptr}; + Tensor _hidden_gate{nullptr}; + Tensor _mm_projection_res{nullptr}; + Tensor _projection_outstage_res{nullptr}; + Tensor _projection_out_res{nullptr}; + Tensor _projection_accumulate_res{nullptr}; + Tensor _ones{nullptr}; std::array<Tensor, _layer_norm_count> _layer_norm_output{}; inline Tensor &get_layer_norm_output(LayerNormGate g) @@ -449,15 +471,15 @@ private: return _layer_norm_output[getGateIndex(g)]; } - bool _is_prepared{ false }; - bool _has_cifg{ false }; - bool _has_cell_clipping{ false }; - bool _has_projection{ false }; - bool _has_projection_clipping{ false }; - bool _has_peephole{ false }; - bool _has_layer_norm{ false }; - bool _projection_tensor_copy_required{ false }; - bool _convert_input_to_forget_weights_to_qsymm8{ false }; + bool _is_prepared{false}; + bool _has_cifg{false}; + bool _has_cell_clipping{false}; + bool _has_projection{false}; + bool _has_projection_clipping{false}; + bool _has_peephole{false}; + bool _has_layer_norm{false}; + bool _projection_tensor_copy_required{false}; + bool _convert_input_to_forget_weights_to_qsymm8{false}; }; } // namespace arm_compute #endif /* ARM_COMPUTE_NEQLSTMLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NERNNLayer.h b/arm_compute/runtime/NEON/functions/NERNNLayer.h index 667d3144ac..af7f464ac9 100644 --- a/arm_compute/runtime/NEON/functions/NERNNLayer.h +++ b/arm_compute/runtime/NEON/functions/NERNNLayer.h @@ -72,7 +72,13 @@ public: * @param[in,out] hidden_state Output tensor of shape [num_units, batch_size]. Data types supported: Same as @p input * @param[in] info Activation layer parameter. */ - void configure(const ITensor *input, const ITensor *weights, const ITensor *recurrent_weights, const ITensor *bias, ITensor *hidden_state, ITensor *output, ActivationLayerInfo &info); + void configure(const ITensor *input, + const ITensor *weights, + const ITensor *recurrent_weights, + const ITensor *bias, + ITensor *hidden_state, + ITensor *output, + ActivationLayerInfo &info); /** Initialize the function * * @param[in] input Input is a 2-D tensor of shape [input_size, batch_size]. Data types supported: F16/F32 @@ -85,7 +91,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *recurrent_weights, const ITensorInfo *bias, const ITensorInfo *hidden_state, const ITensorInfo *output, + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *recurrent_weights, + const ITensorInfo *bias, + const ITensorInfo *hidden_state, + const ITensorInfo *output, const ActivationLayerInfo &info); // Inherited methods overridden: diff --git a/arm_compute/runtime/NEON/functions/NEROIAlignLayer.h b/arm_compute/runtime/NEON/functions/NEROIAlignLayer.h index ea1af4daea..b06ebe899d 100644 --- a/arm_compute/runtime/NEON/functions/NEROIAlignLayer.h +++ b/arm_compute/runtime/NEON/functions/NEROIAlignLayer.h @@ -77,7 +77,10 @@ public: * * @return a Status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *rois, ITensorInfo *output, const ROIPoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *rois, + ITensorInfo *output, + const ROIPoolingLayerInfo &pool_info); }; } // namespace arm_compute #endif /* ARM_COMPUTE_NEROIALIGNLAYER_H */ diff --git a/arm_compute/runtime/NEON/functions/NEROIPoolingLayer.h b/arm_compute/runtime/NEON/functions/NEROIPoolingLayer.h index 2992b3eb95..929111ad4b 100644 --- a/arm_compute/runtime/NEON/functions/NEROIPoolingLayer.h +++ b/arm_compute/runtime/NEON/functions/NEROIPoolingLayer.h @@ -26,6 +26,7 @@ #include "arm_compute/core/IArray.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute @@ -73,7 +74,8 @@ public: * @note The z dimensions of @p output tensor and @p input tensor must be the same. * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. */ - void configure(const ITensor *input, const ITensor *rois, const ITensor *output, const ROIPoolingLayerInfo &pool_info); + void + configure(const ITensor *input, const ITensor *rois, const ITensor *output, const ROIPoolingLayerInfo &pool_info); // Inherited methods overridden: void run() override; @@ -91,7 +93,10 @@ public: * @note The fourth dimension of @p output tensor must be the same as the number of elements in @p rois array. * @return a Status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *rois, const ITensorInfo *output, const ROIPoolingLayerInfo &pool_info); + static Status validate(const ITensorInfo *input, + const ITensorInfo *rois, + const ITensorInfo *output, + const ROIPoolingLayerInfo &pool_info); private: std::unique_ptr<NEROIPoolingLayerKernel> _roi_kernel; diff --git a/arm_compute/runtime/NEON/functions/NERange.h b/arm_compute/runtime/NEON/functions/NERange.h index cb14c8fdde..609456a4ef 100644 --- a/arm_compute/runtime/NEON/functions/NERange.h +++ b/arm_compute/runtime/NEON/functions/NERange.h @@ -26,6 +26,7 @@ #include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" + #include <memory> namespace arm_compute diff --git a/arm_compute/runtime/NEON/functions/NEReduceMean.h b/arm_compute/runtime/NEON/functions/NEReduceMean.h index caaee8284a..5b8d8cdf2b 100644 --- a/arm_compute/runtime/NEON/functions/NEReduceMean.h +++ b/arm_compute/runtime/NEON/functions/NEReduceMean.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NEON_REDUCE_MEAN_H #define ARM_COMPUTE_NEON_REDUCE_MEAN_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/MemoryGroup.h" #include "arm_compute/runtime/NEON/functions/NEReductionOperation.h" #include "arm_compute/runtime/NEON/functions/NEReshapeLayer.h" @@ -81,7 +80,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const Coordinates &reduction_axis, bool keep_dims, const ITensorInfo *output); + static Status + validate(const ITensorInfo *input, const Coordinates &reduction_axis, bool keep_dims, const ITensorInfo *output); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEReductionOperation.h b/arm_compute/runtime/NEON/functions/NEReductionOperation.h index 533c10adcf..f5391a6d0e 100644 --- a/arm_compute/runtime/NEON/functions/NEReductionOperation.h +++ b/arm_compute/runtime/NEON/functions/NEReductionOperation.h @@ -25,9 +25,9 @@ #define ARM_COMPUTE_NEREDUCTIONOPERATION_H #include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/NEON/functions/NEReshapeLayer.h" #include "arm_compute/runtime/Tensor.h" + #include <memory> namespace arm_compute @@ -88,7 +88,11 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, unsigned int axis, ReductionOperation op, bool keep_dims = true); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + unsigned int axis, + ReductionOperation op, + bool keep_dims = true); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NEReorderLayer.h b/arm_compute/runtime/NEON/functions/NEReorderLayer.h index eb777f1925..e3fa7b9c16 100644 --- a/arm_compute/runtime/NEON/functions/NEReorderLayer.h +++ b/arm_compute/runtime/NEON/functions/NEReorderLayer.h @@ -66,7 +66,10 @@ public: * @param[in] input_wf WeightFormat of input. * @param[in] output_wf WeightFormat of output. */ - void configure(const ITensor *input, ITensor *output, arm_compute::WeightFormat input_wf, arm_compute::WeightFormat output_wf); + void configure(const ITensor *input, + ITensor *output, + arm_compute::WeightFormat input_wf, + arm_compute::WeightFormat output_wf); /** Static function to check if given info will lead to a valid configuration of @ref NEReorderLayer * @@ -74,7 +77,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, arm_compute::WeightFormat input_wf, arm_compute::WeightFormat output_wf); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + arm_compute::WeightFormat input_wf, + arm_compute::WeightFormat output_wf); // Inherited methods overridden: void run() override; @@ -85,4 +91,4 @@ private: } // namespace arm_compute #endif /* ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEREORDERLAYER */ -#endif // defined(__aarch64__)
\ No newline at end of file +#endif // defined(__aarch64__) diff --git a/arm_compute/runtime/NEON/functions/NEReverse.h b/arm_compute/runtime/NEON/functions/NEReverse.h index f58eb2373f..e03e415068 100644 --- a/arm_compute/runtime/NEON/functions/NEReverse.h +++ b/arm_compute/runtime/NEON/functions/NEReverse.h @@ -24,9 +24,8 @@ #ifndef ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEREVERSE_H #define ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEREVERSE_H -#include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" namespace arm_compute { @@ -68,7 +67,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const ITensorInfo *axis, const bool use_inverted_axis = false); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const ITensorInfo *axis, + const bool use_inverted_axis = false); }; } // namespace arm_compute #endif // ACL_ARM_COMPUTE_RUNTIME_NEON_FUNCTIONS_NEREVERSE_H diff --git a/arm_compute/runtime/NEON/functions/NEScale.h b/arm_compute/runtime/NEON/functions/NEScale.h index 0920ff3802..72dfa3bda4 100644 --- a/arm_compute/runtime/NEON/functions/NEScale.h +++ b/arm_compute/runtime/NEON/functions/NEScale.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_NESCALEIMAGE_H #define ARM_COMPUTE_NESCALEIMAGE_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/KernelDescriptors.h" #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NESlice.h b/arm_compute/runtime/NEON/functions/NESlice.h index ac79a5c633..70a688d3b0 100644 --- a/arm_compute/runtime/NEON/functions/NESlice.h +++ b/arm_compute/runtime/NEON/functions/NESlice.h @@ -85,7 +85,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); // Inherited methods overridden: void run() override; @@ -129,7 +130,8 @@ public: * * @return A status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); + static Status + validate(const ITensorInfo *input, const ITensorInfo *output, const Coordinates &starts, const Coordinates &ends); }; } // namespace experimental } // namespace arm_compute diff --git a/arm_compute/runtime/NEON/functions/NESpaceToBatchLayer.h b/arm_compute/runtime/NEON/functions/NESpaceToBatchLayer.h index ad8c1467d0..5dee61a4a8 100644 --- a/arm_compute/runtime/NEON/functions/NESpaceToBatchLayer.h +++ b/arm_compute/runtime/NEON/functions/NESpaceToBatchLayer.h @@ -24,9 +24,9 @@ #ifndef ARM_COMPUTE_NESPACETOBATCHLAYER_H #define ARM_COMPUTE_NESPACETOBATCHLAYER_H +#include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" -#include "arm_compute/core/Types.h" #include <memory> namespace arm_compute @@ -82,7 +82,12 @@ public: * @param[in] padding_right The padding at the end of every dimension of the output tensor. * @param[out] output Tensor output. Data types supported: same as @p input */ - void configure(const ITensor *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, ITensor *output); + void configure(const ITensor *input, + const int block_shape_x, + const int block_shape_y, + const Size2D &padding_left, + const Size2D &padding_right, + ITensor *output); /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToBatchLayer * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -92,7 +97,10 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *block_shape, const ITensorInfo *paddings, const ITensorInfo *output); + static Status validate(const ITensorInfo *input, + const ITensorInfo *block_shape, + const ITensorInfo *paddings, + const ITensorInfo *output); /** Static function to check if given info will lead to a valid configuration of @ref NESpaceToBatchLayer (Static block shape and paddings) * * @param[in] input Tensor input info. Supported tensor rank: 4. Data types supported: All. @@ -104,7 +112,12 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const int block_shape_x, const int block_shape_y, const Size2D &padding_left, const Size2D &padding_right, const ITensorInfo *output); + static Status validate(const ITensorInfo *input, + const int block_shape_x, + const int block_shape_y, + const Size2D &padding_left, + const Size2D &padding_right, + const ITensorInfo *output); // Inherited methods overridden: void run() override; diff --git a/arm_compute/runtime/NEON/functions/NESplit.h b/arm_compute/runtime/NEON/functions/NESplit.h index 206f299c06..36358a7094 100644 --- a/arm_compute/runtime/NEON/functions/NESplit.h +++ b/arm_compute/runtime/NEON/functions/NESplit.h @@ -26,7 +26,6 @@ #include "arm_compute/core/ITensor.h" #include "arm_compute/core/Types.h" - #include "arm_compute/runtime/CPP/functions/CPPSplit.h" #include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/NEON/functions/NESlice.h" diff --git a/arm_compute/runtime/NEON/functions/NEStridedSlice.h b/arm_compute/runtime/NEON/functions/NEStridedSlice.h index 4b14d946f6..fa1113ffec 100644 --- a/arm_compute/runtime/NEON/functions/NEStridedSlice.h +++ b/arm_compute/runtime/NEON/functions/NEStridedSlice.h @@ -71,9 +71,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - void configure(const ITensor *input, ITensor *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + void configure(const ITensor *input, + ITensor *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Static function to check if given info will lead to a valid configuration of @ref NEStridedSlice * @@ -89,9 +94,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); // Inherited methods overridden: void run() override; @@ -121,9 +131,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - void configure(const ITensorInfo *input, ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + void configure(const ITensorInfo *input, + ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); /** Static function to check if given info will lead to a valid configuration of @ref NEStridedSlice * @@ -139,9 +154,14 @@ public: * @param[in] shrink_axis_mask (Optional) If the ith bit of shrink_axis_mask is set, it implies that the ith specification shrinks the dimensionality by 1. * A slice of size 1 starting from starts[i] in the dimension must be preserved. */ - static Status validate(const ITensorInfo *input, const ITensorInfo *output, - const Coordinates &starts, const Coordinates &ends, const BiStrides &strides, - int32_t begin_mask = 0, int32_t end_mask = 0, int32_t shrink_axis_mask = 0); + static Status validate(const ITensorInfo *input, + const ITensorInfo *output, + const Coordinates &starts, + const Coordinates &ends, + const BiStrides &strides, + int32_t begin_mask = 0, + int32_t end_mask = 0, + int32_t shrink_axis_mask = 0); }; } // namespace experimental } // namespace arm_compute diff --git a/arm_compute/runtime/NEON/functions/NETile.h b/arm_compute/runtime/NEON/functions/NETile.h index 915e5aa1da..001a0a4128 100644 --- a/arm_compute/runtime/NEON/functions/NETile.h +++ b/arm_compute/runtime/NEON/functions/NETile.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NETILE_H #define ARM_COMPUTE_NETILE_H -#include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/NEON/INESimpleFunctionNoBorder.h" namespace arm_compute { diff --git a/arm_compute/runtime/NEON/functions/NETranspose.h b/arm_compute/runtime/NEON/functions/NETranspose.h index 581fe74309..5d2d1f1b01 100644 --- a/arm_compute/runtime/NEON/functions/NETranspose.h +++ b/arm_compute/runtime/NEON/functions/NETranspose.h @@ -24,9 +24,8 @@ #ifndef ARM_COMPUTE_NETRANSPOSE_H #define ARM_COMPUTE_NETRANSPOSE_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" +#include "arm_compute/runtime/IFunction.h" #include <memory> @@ -83,4 +82,4 @@ private: std::unique_ptr<Impl> _impl; }; } // namespace arm_compute -#endif /* ARM_COMPUTE_NETRANSPOSE_H */
\ No newline at end of file +#endif /* ARM_COMPUTE_NETRANSPOSE_H */ diff --git a/arm_compute/runtime/NEON/functions/NEUnstack.h b/arm_compute/runtime/NEON/functions/NEUnstack.h index 079fee5b9e..e1af96d08d 100644 --- a/arm_compute/runtime/NEON/functions/NEUnstack.h +++ b/arm_compute/runtime/NEON/functions/NEUnstack.h @@ -26,7 +26,6 @@ #include "arm_compute/core/Types.h" #include "arm_compute/runtime/IFunction.h" - #include "arm_compute/runtime/NEON/functions/NEStridedSlice.h" #include <memory> diff --git a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h index f6f0185e7d..7f4e354362 100644 --- a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h @@ -24,10 +24,9 @@ #ifndef ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H #define ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H -#include "arm_compute/runtime/IFunction.h" - #include "arm_compute/core/Types.h" #include "arm_compute/function_info/ActivationLayerInfo.h" +#include "arm_compute/runtime/IFunction.h" #include "arm_compute/runtime/Tensor.h" #include <memory> @@ -87,8 +86,13 @@ public: * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation * available which may introduce a drop of accuracy as well. Default is false */ - void configure(const ITensor *input, const ITensor *weights, const ITensor *biases, ITensor *output, const PadStrideInfo &conv_info, const ActivationLayerInfo &act_info = ActivationLayerInfo(), - bool enable_fast_math = false); + void configure(const ITensor *input, + const ITensor *weights, + const ITensor *biases, + ITensor *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); // Inherited methods overridden: void run() override; @@ -100,8 +104,13 @@ public: * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *biases, const ITensorInfo *output, const PadStrideInfo &conv_info, - const ActivationLayerInfo &act_info = ActivationLayerInfo(), bool enable_fast_math = false); + static Status validate(const ITensorInfo *input, + const ITensorInfo *weights, + const ITensorInfo *biases, + const ITensorInfo *output, + const PadStrideInfo &conv_info, + const ActivationLayerInfo &act_info = ActivationLayerInfo(), + bool enable_fast_math = false); private: struct Impl; diff --git a/arm_compute/runtime/OffsetLifetimeManager.h b/arm_compute/runtime/OffsetLifetimeManager.h index 2eef61a236..13ebb9fbe3 100644 --- a/arm_compute/runtime/OffsetLifetimeManager.h +++ b/arm_compute/runtime/OffsetLifetimeManager.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_OFFSETLIFETIMEMANAGER_H #include "arm_compute/runtime/ISimpleLifetimeManager.h" - #include "arm_compute/runtime/Types.h" #include <map> @@ -62,7 +61,7 @@ public: // Inherited methods overridden: std::unique_ptr<IMemoryPool> create_pool(IAllocator *allocator) override; - MappingType mapping_type() const override; + MappingType mapping_type() const override; private: // Inherited methods overridden: diff --git a/arm_compute/runtime/OffsetMemoryPool.h b/arm_compute/runtime/OffsetMemoryPool.h index a5c363d866..7250194f85 100644 --- a/arm_compute/runtime/OffsetMemoryPool.h +++ b/arm_compute/runtime/OffsetMemoryPool.h @@ -25,7 +25,6 @@ #define ARM_COMPUTE_OFFSETMEMORYPOOL_H #include "arm_compute/runtime/IMemoryPool.h" - #include "arm_compute/runtime/IMemoryRegion.h" #include "arm_compute/runtime/Types.h" @@ -65,8 +64,8 @@ public: const BlobInfo &info() const; // Inherited methods overridden: - void acquire(MemoryMappings &handles) override; - void release(MemoryMappings &handles) override; + void acquire(MemoryMappings &handles) override; + void release(MemoryMappings &handles) override; MappingType mapping_type() const override; std::unique_ptr<IMemoryPool> duplicate() override; diff --git a/arm_compute/runtime/OperatorTensor.h b/arm_compute/runtime/OperatorTensor.h index 92ae01934b..237585bec2 100644 --- a/arm_compute/runtime/OperatorTensor.h +++ b/arm_compute/runtime/OperatorTensor.h @@ -26,8 +26,8 @@ #include "arm_compute/core/ITensor.h" #include "arm_compute/core/TensorInfo.h" -#include "arm_compute/runtime/Types.h" #include "arm_compute/runtime/experimental/Types.h" +#include "arm_compute/runtime/Types.h" #include <cstdint> diff --git a/arm_compute/runtime/PoolManager.h b/arm_compute/runtime/PoolManager.h index cc50fc04a4..6aa6aef6e2 100644 --- a/arm_compute/runtime/PoolManager.h +++ b/arm_compute/runtime/PoolManager.h @@ -24,10 +24,10 @@ #ifndef ARM_COMPUTE_POOLMANAGER_H #define ARM_COMPUTE_POOLMANAGER_H -#include "arm_compute/runtime/IPoolManager.h" - #include "arm_compute/core/Error.h" #include "arm_compute/runtime/IMemoryPool.h" +#include "arm_compute/runtime/IPoolManager.h" + #include "support/Mutex.h" #include "support/Semaphore.h" @@ -53,9 +53,9 @@ public: PoolManager &operator=(PoolManager &&) = delete; // Inherited methods overridden: - IMemoryPool *lock_pool() override; - void unlock_pool(IMemoryPool *pool) override; - void register_pool(std::unique_ptr<IMemoryPool> pool) override; + IMemoryPool *lock_pool() override; + void unlock_pool(IMemoryPool *pool) override; + void register_pool(std::unique_ptr<IMemoryPool> pool) override; std::unique_ptr<IMemoryPool> release_pool() override; void clear_pools() override; size_t num_pools() const override; @@ -66,5 +66,5 @@ private: std::unique_ptr<arm_compute::Semaphore> _sem; /**< Semaphore to control the queues */ mutable arm_compute::Mutex _mtx; /**< Mutex to control access to the queues */ }; -} // arm_compute +} // namespace arm_compute #endif /*ARM_COMPUTE_POOLMANAGER_H */ diff --git a/arm_compute/runtime/RuntimeContext.h b/arm_compute/runtime/RuntimeContext.h index 23bd267375..d64e609196 100644 --- a/arm_compute/runtime/RuntimeContext.h +++ b/arm_compute/runtime/RuntimeContext.h @@ -54,8 +54,8 @@ public: IAssetManager *asset_manager() override; private: - std::unique_ptr<IScheduler> _owned_scheduler{ nullptr }; - IScheduler *_scheduler{ nullptr }; + std::unique_ptr<IScheduler> _owned_scheduler{nullptr}; + IScheduler *_scheduler{nullptr}; }; } // namespace arm_compute #endif /*ARM_COMPUTE_RUNTIME_CONTEXT_H */ diff --git a/arm_compute/runtime/Scheduler.h b/arm_compute/runtime/Scheduler.h index 9e8add1f95..bd29cbb31f 100644 --- a/arm_compute/runtime/Scheduler.h +++ b/arm_compute/runtime/Scheduler.h @@ -74,8 +74,8 @@ public: static bool is_available(Type t); private: - static Type _scheduler_type; - static std::shared_ptr<IScheduler> _custom_scheduler; + static Type _scheduler_type; + static std::shared_ptr<IScheduler> _custom_scheduler; static std::map<Type, std::unique_ptr<IScheduler>> _schedulers; Scheduler(); diff --git a/arm_compute/runtime/SubTensor.h b/arm_compute/runtime/SubTensor.h index 3ca066e1c8..2badb31b26 100644 --- a/arm_compute/runtime/SubTensor.h +++ b/arm_compute/runtime/SubTensor.h @@ -72,5 +72,5 @@ private: ITensor *_parent; mutable SubTensorInfo _info; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_SUBTENSOR_H */ diff --git a/arm_compute/runtime/Tensor.h b/arm_compute/runtime/Tensor.h index 172c8963f0..e71fbd4a96 100644 --- a/arm_compute/runtime/Tensor.h +++ b/arm_compute/runtime/Tensor.h @@ -59,7 +59,7 @@ public: ITensorInfo *info() const override; ITensorInfo *info() override; uint8_t *buffer() const override; - void associate_memory_group(IMemoryGroup *memory_group) override; + void associate_memory_group(IMemoryGroup *memory_group) override; private: mutable TensorAllocator _allocator; /**< Instance of the basic CPU allocator.*/ diff --git a/arm_compute/runtime/TensorAllocator.h b/arm_compute/runtime/TensorAllocator.h index a5e16c4d90..d819931415 100644 --- a/arm_compute/runtime/TensorAllocator.h +++ b/arm_compute/runtime/TensorAllocator.h @@ -24,7 +24,6 @@ #ifndef ARM_COMPUTE_TENSORALLOCATOR_H #define ARM_COMPUTE_TENSORALLOCATOR_H #include "arm_compute/runtime/ITensorAllocator.h" - #include "arm_compute/runtime/Memory.h" #include "arm_compute/runtime/MemoryGroup.h" diff --git a/arm_compute/runtime/common/LSTMParams.h b/arm_compute/runtime/common/LSTMParams.h index aedb9c0d46..6800faf87f 100644 --- a/arm_compute/runtime/common/LSTMParams.h +++ b/arm_compute/runtime/common/LSTMParams.h @@ -79,7 +79,10 @@ public: * * @return Reference to this LSTMParams object */ - LSTMParams &set_cifg_params(const T *input_to_input_weights, const T *recurrent_to_input_weights, T *cell_to_input_weights, const T *input_gate_bias) + LSTMParams &set_cifg_params(const T *input_to_input_weights, + const T *recurrent_to_input_weights, + T *cell_to_input_weights, + const T *input_gate_bias) { _input_to_input_weights = input_to_input_weights; _recurrent_to_input_weights = recurrent_to_input_weights; @@ -125,8 +128,10 @@ public: * * @return Reference to this LSTMParams object */ - LSTMParams &set_layer_normalization_params(T *input_layer_norm_weights, T *forget_layer_norm_weights, - T *cell_layer_norm_weights, T *output_layer_norm_weights) + LSTMParams &set_layer_normalization_params(T *input_layer_norm_weights, + T *forget_layer_norm_weights, + T *cell_layer_norm_weights, + T *output_layer_norm_weights) { _input_layer_norm_weights = input_layer_norm_weights; _forget_layer_norm_weights = forget_layer_norm_weights; @@ -169,7 +174,10 @@ public: * * @return Reference to this LSTMParams object */ - LSTMParams &set_matmul_scale_params(float input_intermediate_scale, float forget_intermediate_scale, float cell_intermediate_scale, float output_intermediate_scale) + LSTMParams &set_matmul_scale_params(float input_intermediate_scale, + float forget_intermediate_scale, + float cell_intermediate_scale, + float output_intermediate_scale) { _input_intermediate_scale = input_intermediate_scale; _forget_intermediate_scale = forget_intermediate_scale; @@ -338,5 +346,5 @@ private: bool _has_cifg_opt; bool _use_layer_norm; }; -} +} // namespace arm_compute #endif /*ARM_COMPUTE_LSTMPARAMS_H */ |