Age | Commit message (Collapse) | Author |
|
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: I7df65ff96ce232dedd80debc34e23a595f99fd06
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Ibdded86713368ecfdf31c4118dfe8a3404d1e3b8
|
|
* Call dedicated MatMul kernel in ACL
* Add int8 tests
* Add int8 to documentation
* Force tensors to be dynamic (nonConst) as per request of ACL
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I7b7ac20deec8637dc46ca990d339d92c4587cbe4
|
|
* Call dedicated MatMul kernel in ACL
* Add int8 tests
* Add int8 to documentation
* Force tensors to be dynamic (nonConst) as per request of ACL
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I992ae9aae1174214607bf29305f21cdeaf3fdc1b
|
|
output size
* Added fix to Classic and Opaque Delegate which now matches
the TfLiteParser.
* Removed uses of single parameter IsValid function in classic
Convolution visit functions.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I5be91ec25c11354a9b8472d0a429e71e02df6f9c
|
|
* If the dimension Size[n] in a Slice is -1 then it should be treated as
"InputShape[n] - Begin[n]" but the Delegate simply cast the Size to
uint and treated it as 4294967295.
* Added the layer name that includes the node index to the Slice to aid
debugging.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I45fa88b24982c3c97f48d0dc05cf7d9bb6db4074
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: Ie02021ac56a512598760e4c6d05ef1a80f4aec8d
|
|
* Intermediate tensors aren't accessible through the new
Opaque interface yet, so we have to cast to TfLiteNode for now.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ifd91131e5d5ff6cc057b80729fea9afa68ed240b
|
|
Classic and Opaque Delegate
* Implemented unsupported operators in Classic Delegate.
* Added unit tests.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib39eeea53c114b15943e8dc2e796ce64c40cb3a5
|
|
* Added implementation for Split / SplitV in Opaque Delegate
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I1ad2925aece7d4e0567e63ecb768348152a70a7a
|
|
* Moved CreateOutputTensorShape function to common DelegateUtils.hpp
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I3d8a9834ecd6b7cda170cce958677a0dde62824a
|
|
* Change alignment in Shape.hpp
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ibc21cf5c56a1ba2daa7507a5d5c2b7311756f17f
|
|
* Added 2 opaque delegate operators and associated test cases
* Removed IsDynamicTensor check from BatchMatMul as covered by IsValid.
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Change-Id: If7c58cb23ae5c5b8a9451dddfd7b6dfcbf248d4c
|
|
DepthToSpace and Tranpose for opaque delegate
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ie0c608f94a76956e9be75f555824cef865cab395
|
|
Dequantize operator
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: I318cb25f526dfe0f7aa6afcf77971afd8d437209
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I25415793497f0ee08d880539e265b133875a20f7
|
|
* Added visit functions for ElementwiseBinary Ops
* Moved MultiLayerFacade.hpp to common directory and updated both delegates to use it
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I84b8bd74d15a194895e63da47c29be994531a889
|
|
Lstm, Pooling2d and Pooling3d operators for Opaque Delegate
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ic9af1c50589285ab359661699d32a889cd267cd9
|
|
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Change-Id: Id4b02b951ed81c69171f6af2d0d327175c6e5d3c
|
|
opaque delegate
* Add check for TF version so tests work with opaque and classic
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: I3a6699150afabfc6200e1cd4d264a1f7c65c5f60
|
|
LogicalBinary operators for opaque delegate
* Report the operator as part of the layer name for:
- LogicalBinary,
- ElementwiseUnary
- Comparison
- Activation
* Fixing indentation in Gather.hpp
* Removing not needed includes in Gather, GatherNd and Comparison
* Correct end of namespace comment in Comparison
* Correct log from TfLiteArmnnDelegate to TfLiteArmnnOpaqueDelegate
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ia0d497709309e912d31eb4b6db0fef9e79b7a3af
|
|
and Prelu operators for Opaque Delegate
* Added missing headers to opaque/CMakeLists.txt (Control and Comparison)
* Cleaned up Control.hpp headers.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I442edb9c467b515b130fbaf02879f0802006255f
|
|
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: Id2b6a4f70b1cb50e5f7f7ab4e30487b3816c9ad4
|
|
* Added Opaque ArgMinMax and associated test cases
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Change-Id: I098b94cc35707370a0bbc7456bfdd48bb47432f0
|
|
Opaque Delegate
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I38304abce1a417bb69aced2a5b38e976ea0cbbc0
|
|
in Opaque Delegate
* Removed input slot check from Connect function as number of TFLite
and Arm NN inputs can differ.
* Moved SetupConcatViewOrigin function to DelegateUtils.hpp
* Simplified validation checks in VistConvolution functions as IsValid
and IsDynamic were already being called.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I858dbe4b643f9d350d9c38ea255ce5effbda4612
|
|
* Added VisitConvolution2d and VisitDepthwiseConv2d functions
to Convolution2d.hpp.
* Enabled Convolution2d and DepthwiseConv2d tests.
* Moved TransposeConvolution2d tests to separate file.
* Added Opaque Delegate shared functions.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ica10c9469fc830f512edad1ad79884f90ae511d0
|
|
* Reference workload
* TfLite Delegate
* TfLite Parser
* Serializer and Deserializer
* Changed fallback tests in delegate to use COS instead of CEIL
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I36e0dbff33694182d1dba0c95d463506428e2f04
|
|
* Adds VisitCast function to Redefine.hpp
* Enables Cast Test for OpaqueUnitTests
* Various Fixes to the opaque delegate to allow operator to run
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I43d42eea5c987d6aed8a0f909a6bf583fddcc94e
|
|
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Change-Id: Id4bdc31e3e6f18ccaef232c29a2d2825c915b21c
|
|
* Added opaque delegate DelegateTestInterpreter implementation
* Moved classic specific tests to ArmnnClassicDelegateTest.cpp
* Moved opaque specific tests to ArmnnOpaqueDelegateTest.cpp
* Removed ArmnnDelegateTest.cpp
* Moved TfLiteStableDelegate implementation to armnn_delegate.cpp
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ifc92b6fb38dc370f3fb88a4daca56d457e74bc2e
|
|
* ABI break on the delegate interface. Bumping the version number.
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Change-Id: I6ef3bc1ea240ef08b67bb3cb9d363a5bbbbdd906
|
|
* Added ArmnnOpaqueDelegate::IdentifyOperatorsToDelegate implementation.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I0b65847358d339a15fc3f729f89deb9b86da0c66
|
|
* Updated all tests to use new DelegateTestInterpreter.
* Fixed some unit tests where the shape was incorrect.
* Add file identifier to FlatBuffersBuilder, as it is required for
validation when creating the model using new API.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I1c4f5464367b35d4528571fa94d14bfaef18fb4d
|
|
* Implement Armnn Subgraph methods in opaque armnn_delegate.cpp
* Temporarily remove opaque delegate from DelegateUnitTests
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I43557eb0b96a22c896890593a3ed9ca9744994bb
|
|
* Implement OpaqueDelegateUtils.hpp using new accessors.
* Moved classic delegate utils to ClassicDelegateUtils.hpp.
* DelegateUtils.hpp now contains common utils.
* Removed unused ConnectConstant function.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I0fa611c82f5c06b0b7e0c37bfc343e09fb6a96c9
|
|
* Also added cmake for the new layers to reduce merge conflicts.
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: Ieb59aa2b7e2a18c57c9357b8d5b5cd63d8211c85
|
|
* New folders created:
* common is for common code where TfLite API is not used
* classic is for existing delegate implementations
* opaque is for new opaque delegate implementation,
* tests is for shared between existing Delegate and Opaque Delegate which have test utils to work which delegate to use.
* Existing delegate is built to libarmnnDelegate.so and opaque delegate is built as libarmnnOpaqueDelegate.so
* Opaque structure is introduced but no API is added yet.
* CmakeList.txt and delegate/CMakeList.txt have been modified and 2 new CmakeList.txt added
* Rename BUILD_ARMNN_TFLITE_DELEGATE as BUILD_CLASSIC_DELEGATE
* Rename BUILD_ARMNN_TFLITE_OPAQUE_DELEGATE as BUILD_OPAQUE_DELEGATE
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ib682b9ad0ac8d8acdc4ec6d9099bb0008a9fe8ed
|
|
When creating a flatbuffers model, we need to provide an empty buffer 0 that is
reserved by tensorflow. When creating empty buffers for inputs and outputs we
can not pass in an empty vector, or tflite will assume that we know how many bytes to
allocate in advance. Instead we need to only pass in the builder.
* Update libraries in FindTfLite.cmake
* Add nullptr to delegate struct for OpaqueDelegateBuilder
* Fix issue in unit tests where Flatbuffers model was not being parsed by tflite
* Tensorflow 2.12 now includes C++ 17 features. Update our cmake build
to require a compiler to support these features.
* Change minimum cmake in Arm NN to 3.7 as that's the minimum for the
delegate build.
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I7d15b196b8c59b1914f8fc1c4c2f8960630c069c
|
|
!android-nn-driver:9329
* Added ElementwiseBinaryLayer that can represent all ElementwiseBinary
operations including Add, Div, Sub, Maximum, Mul and Minimum.
* Updated Delegate to use ElementwiseBinaryLayer instead of the Add,
Div, Sub, Maximum, Mul and Minimum layers.
* Updated Deserializer to use ElementwiseBinaryLayer instead of the Add,
Div, Sub, Maximum, Mul and Minimum layers.
* Updated OnnxParser to use ElementwiseBinaryLayer instead of the Add
layer.
* Updated TfLiteParser to use ElementwiseBinaryLayer instead of the Add,
Div, Sub, Maximum, Mul and Minimum layers.
* Updated CL and Neon tests to use ElementwiseBinaryLayer.
* Updated CL and Neon Backend Specific Optimizations to accept
ElementBinaryLayers as well as Add, Div, Mul, Sub, Maximum and Minimum
layers.
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I7cbb96b60eb01f0e2b57b0541016d48a08b86c75
|
|
This makes it easier to use, particularly in backends where it
is common to pass around const pointers to IConnectableLayer.
The non-constant version is rewritten to use the constant version.
Signed-off-by: Matthew Bentham <matthew.bentham@arm.com>
Change-Id: Id3a8384447e93c213299a85ade9a667df5960534
|
|
Previously we were adding a reshape layer to "broadcast" tensors
for elementwise operations. This broadcast was happening too late
and was really just an expand dims. This was breaking the constant
attributes of tensors and layer support of certain backends.
* Remove addition of reshape layer when expanding dimensions
* Replace broadcast function with expand dims to equal rank function
* Fix some error status checks in various layers
* Add new TensorUtil function that expands dims to a defined rank
* Add unit tests to new TensorUtil function
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I31aca47c98075fef4f86864a15470f5faa55ab8d
|
|
ProfilingOptions is not used in DelegateOptions. Instead the parameters
are passed in through the RuntimeOptions. This is done in ExecuteNetwork
and TfliteExecutor.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: Iaab3d4ef277c47e1ff82a51ba2648f5f51ec3e2c
|
|
In the TLCT tests we were failing many tests because they used
constant tensors as data input for the layers. We had the functionality
in place but we didnt have it spread across the visit functions.
* Check if inputs are constant tensors and attempt to assign them
to input slot of layers.
* Add missing checks to some functions that return a kTfLiteStatus
so we can see if they fail
* Clean up CreateConstTensor function
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I8610b770aea56932a98f91c961d59b3de47c2ab5
|
|
* Add initialization of activation type in delegate layers with
fused activations
* Remove unused variable activationOutputInfo on layers with
fused activations
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: Ieb2ba00516a159639871c391069faf162cbb666d
|
|
In layers that support fused activations, we check for activation
layer support after we already create the base layer. This breaks
the fallback as we already added the base layer to the graph.
* Creates ValidateFusedActivation shared function
* Moves Activation validation higher in the VisitFunction
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I239af360923f695fc374ddeaeefa24c062eaf9e8
|
|
* When the tfLiteExecutor attempts to populate the input tensors it did
not check whether the tensor was constant. This was causing
segmentation faults.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I80a4cc788de4ffe08afb2df9185d04fcb8b27c3a
|
|
* TFLite Parser:
* Fixed issue in ParseReshape where the targetShape wasn't always calculated correctly
* Fixed issue in ParseFullyConnected where the wrong name was used for the ReshapeLayer
* Added an ExpandDims to the FullyConnected to ensure that we reshape the output correctly
* TFLite Delegate:
* Added an ExpandDims to the FullyConnected to ensure that we reshape the output correctly
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I129dfcb8543f8a3a297c0589c841be20ef3b6407
|
|
* Added ability to reduce dimension sizes when calling BuildArmComputeTensorInfo or
BuildArmComputeTensorShapes, this will attempt to remove leading 1s in order to
squeeze the number of dimensions but retain the size.
* Changed ClBatchMatMulWorkload to attempt to squeeze the number of dimensions to 3
as the CL Gemm Kernel can only support up to 3 dimensions.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I6b3d0886c5b97fdb686838fc3dc292833ddc4643
|
|
one works fine
* All ArmNNExecutors now share a single IRuntime.
* All armnn_delegates now share a single IRuntime.
* Increased delegate major version.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I95cbdc32655ec0beb476dbb2d60f1a0209df8f04
|