diff options
author | Manuel Bottini <manuel.bottini@arm.com> | 2020-04-08 10:15:51 +0100 |
---|---|---|
committer | Manuel Bottini <manuel.bottini@arm.com> | 2020-04-23 17:53:59 +0000 |
commit | 2b84be544e4a27f7e8e80827e9c85c8f0d58b4ce (patch) | |
tree | 078051a911f9b8883a3f11955cfd3b7ba0d7d9f3 /arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h | |
parent | 0de45d0a8009e19331c4e29d617fa183167c513a (diff) | |
download | ComputeLibrary-2b84be544e4a27f7e8e80827e9c85c8f0d58b4ce.tar.gz |
COMPMID-3280: Make all ML primitives for CL use the new interface - Part 2
- CLFunctions have been updated
Change-Id: Ie3256a6c775bc12f3126482bd8e8a46da54b267c
Signed-off-by: Manuel Bottini <manuel.bottini@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3053
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h')
-rw-r--r-- | arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h | 51 |
1 files changed, 48 insertions, 3 deletions
diff --git a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h index 4952029c9d..6d1181eefe 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMConvolutionLayer.h @@ -62,6 +62,16 @@ public: * @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 *weights, const ICLTensor *biases, ICLTensor *output, unsigned int num_groups = 1); + /** Set the input and output tensors. + * + * @param[in] compile_context The compile context to be used. + * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. + * Data type supported: QASYMM8/QASYMM8_SIGNED/QSYMM8_PER_CHANNEL/F16/F32. + * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p weights. + * @param[out] output Destination tensor. Data types supported: Same as @p weights. + * @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 *weights, const ICLTensor *biases, ICLTensor *output, unsigned int num_groups = 1); /** Static function to check if given info will lead to a valid configuration of @ref CLConvolutionLayerReshapeWeights * * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. @@ -94,9 +104,20 @@ public: */ void configure(const ICLTensor *input, const ICLTensor *biases, unsigned int num_groups) { + configure(CLKernelLibrary::get().get_compile_context(), input, biases, num_groups); + } + /** Configures the @ref CLConvolutionLayerReshapeWeights function + * + * @param[in] compile_context The compile context to be used. + * @param[in] input Input tensor. Data type supported: QASYMM8/QASYMM8_SIGNED/F16/F32. + * @param[in] biases Biases tensor. Data type supported: Same as @p input. + * @param[in] num_groups Number of groups when performing a grouped convolution. + */ + void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *biases, unsigned int num_groups) + { _bias_bit = (biases != nullptr) ? 1 : 0; _num_groups = num_groups; - _func.configure(input, biases, &_output, num_groups); + _func.configure(compile_context, input, biases, &_output, num_groups); } //Inherited method override @@ -178,6 +199,28 @@ public: */ 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. + * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], + * while every optional dimension from 4 and above represent a batch of inputs. + * Data types supported: QASYMM8/F16/F32. + * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. + * Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8. + * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. + * Data type supported: Should match @p input data type, except for input of QASYMM8 type where biases should be of S32 type. + * @param[out] output Destination tensor. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs. + * Data types supported: Same as @p input. + * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. + * @param[in] weights_info Specifies if the weights tensor has been reshaped with CLWeightsReshapeKernel. If this is not part of the fully connected layer the weights + * tensor has also been transposed with CLGEMMReshapeRHSMatrixKernel. Data type supported: Same as @p input. + * @param[in] dilation (Optional) Dilation, in elements, across x and y. Defaults to (1, 1). + * @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); /** 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], @@ -208,6 +251,7 @@ public: private: /** Configures the appropriate matrix multiply routine * + * @param[in] compile_context The compile context to be used. * @param[in] input Input tensor. Data types supported: QASYMM8/F16/F32. * @param[in] weights Weights tensor. Data type supported: Same as @p input or QASYMM8/QSYMM8_PER_CHANNEL when @p input is QASYMM8. * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. @@ -218,8 +262,9 @@ private: * @param[in] gemm_3d_depth Depth of GEMM 3D * @param[in] act_info Activation to apply after the matrix multiplication */ - void configure_mm(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, const GEMMLowpOutputStageInfo &gemmlowp_output_stage, int gemm_3d_depth, - const ActivationLayerInfo &act_info); + void configure_mm(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *weights, const ICLTensor *biases, ICLTensor *output, + const GEMMLowpOutputStageInfo &gemmlowp_output_stage, + int gemm_3d_depth, const ActivationLayerInfo &act_info); /** Static function to check if given info will lead to a valid configuration of @ref CLGEMMConvolutionLayer matrix multiply routines * * @param[in] input Input tensor info. Data types supported: QASYMM8/F16/F32. |