aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2023-05-08IVGCVSW-7626 Add Execute Network for Opaque DelegateNarumol Prangnawarat
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com> Change-Id: Ibdded86713368ecfdf31c4118dfe8a3404d1e3b8
2023-05-08IVGCVSW-7308 Add GpuAcc Batch MatMul workloadTeresa Charlin
* 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
2023-05-08IVGCVSW-7307 Add CpuAcc Batch MatMul WorkloadTeresa Charlin
* 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
2023-05-08BugFix: Calculate explicit padding for Delegate Transpose Convolution using ↵Matthew Sloyan
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
2023-05-06Update ACL pin to 23.05 release branchNikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Iadba213e78f4c7eed7d1eea2206dcd1d977446b5
2023-05-05MLCE-1050 Error handing Slice operatorsMike Kelly
* 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
2023-05-05IVGCVSW-7423 Add ArmnnDelegatePluginNarumol Prangnawarat
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com> Signed-off-by: Ryan OShea <ryan.oshea3@arm.com> Change-Id: Ie02021ac56a512598760e4c6d05ef1a80f4aec8d
2023-05-05IVGCVSW-7618 Implement UnidirectionalSequenceLstm operator for Opaque DelegateMatthew Sloyan
* 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
2023-05-05IVGCVSW-7484 Update versions in Arm NN for 23.05Nikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Ib0d7835355e82d94b66b27a25b56b788c4606ae8
2023-05-05Update ACL pin to 352c07ddd49842b5c3a8e5a2b5a90832bfb70091Nikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Iaf3242b4d06bd043fe25a00722da6cb6406d013b
2023-05-04IVGCVSW-7605 IVGCVSW-7604 Implement Squeeze and ExpandDims operators for ↵Matthew Sloyan
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
2023-05-04IVGCVSW-7612 Implement Split OperatorsDavid Monahan
* Added implementation for Split / SplitV in Opaque Delegate Signed-off-by: David Monahan <david.monahan@arm.com> Change-Id: I1ad2925aece7d4e0567e63ecb768348152a70a7a
2023-05-04Add tflite-opaque-delegate flag in build-tool validation scriptNikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I402a4c74610b5dc8a3ca6e22218b58fbe59474ed
2023-05-04Update ACL pin to 57132943e0df00aa008b90614ea5a9fa8b2dc18aTeresa Charlin
# Fix im2col for fast-maths mode with padding. Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I6140f5435b81598d29a30f1f6747fd59e2304101
2023-05-03Minor fixes to the build android ndk guide scriptNikhil Raj
* Aligns both sh script and md file * Change default branch to main as 'latest' release will change every 3 month * Clones github repos instead of mlplatform repos Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I4bed6c4c231b5f3bab0abf92b0febf825ddb4ce7
2023-05-03Fix bug in the build-armnn.sh in build toolNikhil Raj
* flag should be flag_tflite_classic_delegate instead of flag_tflite_delegate Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I4c1f6406d88ee91c032f8edcf1196f2e8e8bcb92
2023-05-03IVGCVSW-7528 Update Arm NN documentation to reflect use of Tensorflow 2.12Nikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Ib3aaba6d2bbe52978fc3e8873c78e4c2989131ed
2023-05-03Update ACL pin to 60ab4e66ea3cb85042035fd1aafbfea666bb4ea7Teresa Charlin
* Fix export_to_cl_image issue in the fp16 GeMM implementation Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I07440e48243abb599a0bd9e6d96d8c3291e87e91
2023-05-02IVGCVSW-7404 Replacing asserts with exceptions in CopyTensorContentsGenericColm Donelan
* The source and destination size checks in CopyTensorContentsGeneric are handled by asserts instead of exceptions. * Adding unit tests. Signed-off-by: Colm Donelan <colm.donelan@arm.com> Change-Id: Ia00c07158afde6768002dc6059067fd08e47fcff
2023-05-02IVGCVSW-7603 Implement Reshape operators for Opaque DelegateMatthew Sloyan
* Moved CreateOutputTensorShape function to common DelegateUtils.hpp Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com> Change-Id: I3d8a9834ecd6b7cda170cce958677a0dde62824a
2023-05-02Update ACL pin to f0ff76dbfc9137d0dfc5e99666e24c7a2ca8b072Nikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I5d8b9e38b50ee56e03c8421e4ddb9a02eade5d47
2023-05-02IVGCVSW-7536 Update Build Tool for Opaque DelegateNikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Iaf3f1b5e4e41d0d946a21504c6d44e47d83273e0
2023-04-28IVGCVSW-7610 IVGCVSW-7616 Slice and StridedSlice for opaque delegateTeresa Charlin
* Change alignment in Shape.hpp Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: Ibc21cf5c56a1ba2daa7507a5d5c2b7311756f17f
2023-04-28IVGCVSW-7606 IVGCVSW-7607 Add Resize and Reduce to Opaque DelegateJohn Mcloughlin
* 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
2023-04-28IVGCVSW-7611 IVGCVSW-7614 IVGCVSW-7615 IVGCVSW-7617 Softmax, SpaceToDepth, ↵Teresa Charlin
DepthToSpace and Tranpose for opaque delegate Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: Ie0c608f94a76956e9be75f555824cef865cab395
2023-04-28IVGCVSW-7602 IVGCVSW-7602 Implement opaque delegate for Quantize + ↵Francis Murtagh
Dequantize operator Signed-off-by: Francis Murtagh <francis.murtagh@arm.com> Change-Id: I318cb25f526dfe0f7aa6afcf77971afd8d437209
2023-04-28IVGCVSW-7596 IVGCVSW-7619 IVGCVSW-7597 Pack, Unpack and Pad for opaque delegateTeresa Charlin
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I25415793497f0ee08d880539e265b133875a20f7
2023-04-28Update ACL pin to 99b6fd210d84511b5d746871b8510a482a2950b8Nikhil Raj
* Fix filelist issue with CPU MatMul Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Iaeaae6467df23b36e48fe6aa707d61c28515ad16
2023-04-28Make Convert workloads use arm_compute::NECast in CpuAcc backendMatthew Bentham
NECast can use conversion instructions where they are available so this should in general be faster. Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com> Change-Id: I3f259e17b280a4f4c36f363965ffbc8ee8c4c29f
2023-04-28IVGCVSW-7588 Implement ElementWiseBinary Op for Opaque DelegateDavid Monahan
* 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
2023-04-28IVGCVSW-7608 IVGCVSW-7594 IVGCVSW-7598 IVGCVSW-7599 Implement Floor,Matthew Sloyan
Lstm, Pooling2d and Pooling3d operators for Opaque Delegate Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com> Change-Id: Ic9af1c50589285ab359661699d32a889cd267cd9
2023-04-27IVGCVSW-7576 IVGCVSW-7609 Add BatchMatMul and Shape to Opaque DelegateJohn Mcloughlin
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com> Change-Id: Id4b02b951ed81c69171f6af2d0d327175c6e5d3c
2023-04-27IVGCVSW-7584 IVGCVSW-7585 Implement Conv3d and TransposeConv2d operators for ↵Francis Murtagh
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
2023-04-27Add unit test for Neon Convert workloadsMatthew Bentham
Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com> Change-Id: I501a3e01932d44eca796e93a9383378dafc758c5
2023-04-27IVGCVSW-7589 IVGCVSW-7595 IVGCVSW-7593 ElementwiseUnary, Normalization and ↵Teresa Charlin
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
2023-04-27Update ACL pin to f16eed979ecaa234b308c8eb145c5f9512673a54Nikhil Raj
* Change fp16 GeMM heuristic for Arm® Mali™-G77 Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I7b39b20f9d428750e05333e5a8f4aec49353c5f9
2023-04-27IVGCVSW-7574 IVGCVSW-7590 IVGCVSW-7600 Implement Activation, FullyConnected ↵Matthew Sloyan
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
2023-04-27IVGCVSW-7687 Update TF SHA in build-toolNikhil Raj
* Updating TF to the 2.12 SHA Arm NN will be using * Turn Off JNI build Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: Ib75e16687ac3a2dd4a77ed68f97bf14fdd86eec0
2023-04-27IVGCVSW-7591 IVGCVSW-7592 Add Gather and GatherNd to Opaque DelegateKevin May
Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: Id2b6a4f70b1cb50e5f7f7ab4e30487b3816c9ad4
2023-04-26IVGCVSW-7575 Implement opaque delegate for ArgMinMax operatorJohn Mcloughlin
* Added Opaque ArgMinMax and associated test cases Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com> Change-Id: I098b94cc35707370a0bbc7456bfdd48bb47432f0
2023-04-26IVGCVSW-7577, IVGCVSW-7578 Implement BatchToSpaceNd and SpaceToBatchNd in ↵Kevin May
Opaque Delegate Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: I38304abce1a417bb69aced2a5b38e976ea0cbbc0
2023-04-26BugFix: calculate explicit padding for Transpose Convolution using output sizeTeresa Charlin
* If the output shape is given in Transpose convolution, use it to calculate the padding Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I0bf3dee94c2ce606ed67fb385018b220188c3017
2023-04-26Update cxxopts from 3.0 to 3.1.1Jim Flynn
!android-nn-driver:9431 Signed-off-by: Jim Flynn <jim.flynn@arm.com> Change-Id: I58143445b5c5cf2aafd0838156c9543adce21e6a
2023-04-26IVGCVSW-7579 IVGCVSW-7581 IVGCVSW-7583 Implement Comparison, Concat and Mean ↵Matthew Sloyan
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
2023-04-25Update ACL pin to 05a65e3f397946cacf5c17d8c528a9fad3f1b322Nikhil Raj
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I94a95c84241e6beac63e4d4fc91c1d42a6ed2a5f
2023-04-24IVGCVSW-7584 Implement Conv2d and DepthwiseConv2d operators for Opaque DelegateMatthew Sloyan
* 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
2023-04-21Update ACL pin to 467daef993fe29cc4319058200b7ad797398e4b0Nikhil Raj
* Implement CL kernel for a native batched matmul Quantized - LHS transposed, RHS transposed Signed-off-by: Nikhil Raj <nikhil.raj@arm.com> Change-Id: I00d1499033f7b6279d489c26c17200860daaf299
2023-04-20IVGCVSW-7687 Adding Tf 2.12 build flags to build-tool.Colm Donelan
Signed-off-by: Colm Donelan <colm.donelan@arm.com> Change-Id: I637fb48e5ae0e2bc90c2900b5b35860e0fbcfb25
2023-04-20GitHub #723 Fix missing includeTeresa Charlin
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: I8a0f60f8da878faac8254ecff112c66a92c4888e
2023-04-20IVGCVSW-7685 PyArmNN test failures in nightlyTeresa Charlin
* This was originated when setting scale to 1 in quantization parameters Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> Change-Id: Ic3192a6985f3afc595952eb5bfdaa9b628bbb88d
generated by cgit v1.2.1 (git 2.23.0) at 2024-07-08 16:18:14 +0000