Age | Commit message (Collapse) | Author |
|
* Fix nightly issue caused by gemm_reshaped_only_rhs_mmul kernel
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Iff1edca64a000387c33f12756f15b1ef2fe575cd
|
|
valgrind
* Add end to end unit test to CpuRef, CpuAcc and GpuAcc backends
Resolves: IVGCVSW-8193
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I7be226f084ec814ac72c2c9b3c47c07b3baf0aa5
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I86deb3bb2a587e59f10cfc3efce7118c2156c169
|
|
* Adding a one to many tosa mapping for Quantize
* Added tests
* Resolves IVGCVSW-7175
Signed-off-by: John Mcloughlin <john.mcloughlin@arm.com>
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ia0852fefb618b4a29c2601b9de8b6b2731229801
|
|
* Relates to IVGCVSW-8193 and IVGCVSW-7346
Change-Id: Ieccee93672a5c73297c4ce69d1eaec588e858df0
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I44df03acd348532a54b66541d91610d382a222b7
|
|
* Remove reshape on ClBackend
* Remove unnecessary restriction on NeonBackend remove Reshape
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I79940c9f8609d19b79f2fbe55225ffc8f0d90c25
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ib3b3b450197bb08e0f5879be8c7a129f44a49d06
|
|
The intent of this change is to remove the per backend test cases in
the delegate unit tests. They will be replaced by using DocTest
SUBCASES. The sub cases are paramaterized by the available backends.
The list of available backends are determined by the compilation flags.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: Ia377c7a7399d0e30dc287d7217b3e3b52e1ea074
|
|
* Added ElementwiseBinary EndToEnd tests with inputs of the same shape to avoid Reshape
* Added Slice EndToEnd tests with 4D tensors
* Added TosaReference support for Maximum and TosaRefEndToEnd tests
Signed-off-by: Tianle Cheng <tianle.cheng@arm.com>
Change-Id: I4fa24435a75559e00b110d0e542b4f2bf07b21b4
|
|
* Resolves IVGCVSW-7918
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: Idd15ce139f55895957378f9a9d1471e3e48989bb
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: If300c58fd20baa8aaf0c8587d5312ad71c6f3e13
|
|
* Resolves IVGCVSW-7918
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: Ic2afaa55f7ee88ce4c9b8ea696eef5f28663f8c6
|
|
* Add support for quantized data in TosaRefPreCompiledWorkloadGetOutput.
* Remove extra includes from all TOSA operators headers.
* Added positive and negative unit tests for resize.
* Resolves: IVGCVSW-7346
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ib6e30d018a7a1bf26b380fc794560aae108b26c3
|
|
with "--tflite-executor parser" option
* Allow serialize to dot for the parser delegate
Signed-off-by: Tracy Narine <tracy.narine@arm.com>
Change-Id: Id9edef143f0e0cc02dd7b1bfe87d36035bf44c6d
|
|
* All of the unit tests were including the TfL schema even though it
is not used. This prevented building with delegate and not TfL parser.
* Fixing some template compile errors that were introduced when
FileComparisonExecutorTests.cpp was compiled without TfL parser.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I86742d57a6520ab6bd8336a683fee29de326fff5
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Iaf47b33834d9a4457fb28027d9c28d32c788785d
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ie97e8bbfb16793e8e509211e4b4e73f3ec5a1006
|
|
* Added in missing include MemoryGroup.h
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I2a5f4f25ec8cdd63d90399c042d46e26b3e2a364
|
|
Squashed commit of the following:
IVGCVSW-7159 Add GpuFsa backend skeleton
IVGCVSW-7380 Update the GpuFsa Skeleton to build and load ACL
IVGCVSW-7381 Add IsLayerSupported implementation to GpuFsa backend
IVGCVSW-7382 Implementation of Conv2d within GpuFsa
Signed-off-by: James Conroy <james.conroy@arm.com>
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Id23d9ee598535de7b38a99ca223cdf0ad2102cef
|
|
* Restored previous functionality for compilers with a version
of the C++ standard earlier than C++17
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I0951fc1cac366e74066b6ababa675e19cfb6536b
|
|
* The compiler shipped with NDK r26 has stricter rules around certain
warnings and deprecation notices.
* Fixed warnings for unqualified call to 'std::move'
* Fixed error where the half values weren't being cast to a float
when calling 'std::nan'
* Removed unnecessary subtensor unit tests for neon
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I4ceb46e55ff5f2a754452e3a43de2188d58bf927
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I17b81b4e9a0cc467dacfd2745a34872b58a622a4
|
|
* Add Android compiler flag
* Install llvm package for docker compilation
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I9054b64e372fe9d9181d9b2321be86b3f3d95267
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I53e53ad2d03556327c60730ae669f61f400208f4
|
|
* Adding clock times before and after inferences which SA team requested
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: Idc37085a25ce573cb74340f1be7313a616f561e8
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ia6af7e488e2be3213a918e2fa4c2527f939068cc
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ic69532e6d9101c76a45898e8204aa1ed9bf97e91
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I16efef72b2234ba689749bbd72136b89ef628f26
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ieaa7470f9f19543962e5b121c15e88bd37ca7f3c
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I1cdbe17bdbfa48bc0e35a14eb642d494291f0013
|
|
* Removes warning we had given to users last release about Android 10 (Q) testing
* Reverts the patch where we were checking out main branch of Arm NN instead of
latest release branch in Buildtool
* Updated Android API level in the bianry table in readme
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I1e497b7d0ecf8fa7d9dd7262ca740052c0db444e
|
|
* This is part of IVGCVSW-7938 Update 23.11 Doxygen Documents
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I669e632c95f5be1ac56b407fd1fb252a8597e015
|
|
23.11
* GELU added to activation operators, despite not being part of Android NDK
* BROADCAST_TO added to layers, despite not being implemented in any backend
* CAST SIGNED64 added
Change-Id: I32edb9939d28cb997432762e8ea5904f92c12687
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I6a27ebc1233d77522c1bed1c7e89df95e4d55177
|
|
* Opaque delegate was broken by https://review.mlplatform.org/c/ml/armnn/+/10493/
* Updated ArmNN minor version in pyarmnn
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I9eb3a2901c9383b95079b092e05373680781b1ab
|
|
* Fixed issue where backends were asked to support FP16 layers that would
be optimized out.
* Fixed issue where backends were asked to support non-constant filter
and bias tensors when those tensors would be replaced by constant
tensors during optimization.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Ib54b9cb99d5014e27172841a665daf57d1d5b23d
|
|
* Added ReverseV2 to CL and Neon backends
* Added Cl and Neon ReverseV2 Layer unit tests
Signed-off-by: Tianle Cheng <tianle.cheng@arm.com>
Change-Id: I646275c629caf17dac1950b0cd7083f23f87f387
|
|
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I902815f72242a795ad16e869c10f6370e9675779
|
|
* Updated the exception catch when calling EnqueueWorkload to also print the exception to std::out when caught. Helps greatly with debugging
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I8e9dd4d69986543ff33cd91703bb381d4d062b22
|
|
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: If156b975d37868db77d7c6bb75d884652278e02a
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I37e6baeae214b52365db3c4963ecdca3e29eda25
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ic3c42db6fd87b379b42d610e0d0f56f55580268d
|
|
This script will attempt to execute a tflite model through all available
backends reporting performance and accuracy checks.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: Id60626eddc48b48c497a7f52f3fbd10aa036d997
|
|
* Following resolution of COMPMID-6397 we will allow non
const bias CONV2D to be validated by ACL.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I31c34f3c44fe96885077b3e266f840e51ecf02ec
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I80c9b3e76a04102e13aa77998102fd521c5f0fd5
|
|
* This helps debug when running the DTS
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I623001529e8b56f654e0d93cbb96d3f5b38a7882
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Icde36ce1cc41be753cec7280ccb36293586bdf73
|
|
* Remove mention of "isDepthwise" variable name when not needed and therefore IgnoreUnused call
* Improve error messages and change them to throws in Encoder and Decoder
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I8ce30b5075e1e47d54abc12390265ba8e9ee1405
|
|
* Update ACL pin
* ACL removed softmax_layer_max_shift_exp_sum_quantized_serial cl kernel
* Replace softmax_layer_max_shift_exp_sum_quantized_serial with softmax_x
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I0830605d72999d5a80f06e1b8498bc7b674195e3
|
|
Signed-off-by: Ciara Sookarry <ciara.sookarry@arm.com>
Change-Id: I16ef18a4d4afce0ae17c98cfb5267f51b7708539
|
|
* Add check to disable dynamic bias with quantized datatypes in Conv2D layer
Change-Id: Ie97ebfc2beaeb7fb4325e50c10ae595159346be3
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Iad6d293db59d065d00c9a078456ba0fdadc39776
|