Age | Commit message (Collapse) | Author |
|
* Added TosaRefBackend::OptimizeSubgraphView implementation.
* Generalised TosaRefLayerSupport::IsLayerSupported to work with any
operator.
* Changed TosaCommon.hpp utils to inline functions.
* Added source files for TosaMappings.hpp and AdditionOperator.hpp.
* Fixed multiple defines issue with HALF_ROUND_STYLE and
HALF_ROUND_TIES_TO_EVEN.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib2576ec3fb97faa3a2256b2fb93ec16ac8745760
|
|
* Build Flatc only on x86_64 when cross compiling
* Add Linux CMAKE System name
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: Ibd0f636b3cc149e0492d66330baf0e3707a9d3f7
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I6a541db9a602609282cc6f33af930ca141b83c41
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I6338fe09067a90ef2f52f252b3d5cf2c8810f3ff
|
|
* In TF what ArmNN calls quantized data types can be non-quantized as well.
* This patch creates 2 models:
* ArmNN: model where int8 and uint8 will always be quantized, but scale can be 1 and offset 0
* TFLite: model where int8 and uint8 can be quantized and non-quantized
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Id960f2f30988f2bbec88cb4e0c52c189ac957bae
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I2def6995f81d33e68f1ea45d8d19a1e6294049b1
|
|
* The intention is to keep the flexibility given by the ExNet before the refactor.
* When iteration > inputFiles, we repeat the usage in order
* When iteration < inputFiles, we just discard extra files.
Signed-off-by: Adam Jalkemo <adam.jalkemo@arm.com>
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I2fbe69f8affe0e3a5cc86fc1748164967f0c2d64
|
|
not match"
This reverts commit 6c95836e894f88c4bab6b22f974341f0dd2dddaa.
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I8be2147feb557a0849de5785fb63b464abc7dbb9
|
|
* half.hpp is now required as an include for numpy_utils.h
* DType_FLOAT was changed to DType_FP32.
* DType_FP16 support has been added to ArmNNToDType utility function.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib25be9a5abfbd9db2a90b3e42d592259a6df0e01
|
|
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ie9aa1aa20a5bc837e005a7a36d4d07b4cd8d021b
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I9bc849ec3375c74647b637758e38fdb1870832b7
|
|
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: Ib30fc633a10b6ff8090b50314278fe5dc46fb250
|
|
* Upgrade FlatBuffers to 2.0.6.
* Upgrade TensorFlow to 2.10.0.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I537141ccfea208e2ff5c16d76f288f477efd89f9
|
|
* Currently, the build-tool requires the user provide both
Arm NN and ACL repos, if providing custom repos.
* This change allows the user to provide only a custom
Arm NN repo, making a custom ACL optional in the case of
--neon-backend or --cl-backend not being chosen in BUILD_ARGS.
* Updated script help/usage and docs to reflect this.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I821f07dcd4c32ada60cd33713234b47c330d217e
|
|
* Modify the order of Tf absl libraries in delegate build.
* Disable failing delegate unit tests.
* Add -Wno-comment to delegate ExecuteNetwork build.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I9e4696e032a5e26df8ba578b980456cbac4d0ab0
|
|
* Modify get_tensorflow.sh to point to the 2.10 Tensorflow tag.
* Modify FindTfLite.cmake to handle the break up of libruy.a
* Modify armnn_delegate.cpp to add registration_external field.
* Prevent the serializer/deserializer from using a system installed
version of flatc.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I688f33f387924ba29bf400f60e56f73b2329fdc1
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I07fc2fc524452ff3631abe8aafd5993cb23a2258
|
|
* Add flags -Wno-c99-designator and -Wno-unused-but-set-variable
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I7b1a1a353a0474d7ef967f6f3c8d13c36fa31844
|
|
* Bug fix where files were being overwritten at each debug layer
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I609fdc82afcee925824efb02183c7dbc942fced0
|
|
* Add functionality to print output tensors to file in tempdir
* UnitTests
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: Idfb4c186544187db1fecdfca11c662540f645439
|
|
Change-Id: I594ded82493e9cc48cafa6f00d63769fefba5afe
|
|
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Change-Id: I3573078206272c3a72a2b3acf8781ab458ea6c90
|
|
* Add scons option to generate Map files.
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ic1485f4f544ae1139e2475b95bc6d4f0d4791f02
|
|
* When a model with multiple outputs was used and output to file, e.g.
with "-w ./boxes,./classes,./scores,./detection", the results where
not saved in the correct files.
* Applies only to the ArmNNExecutor.
Change-Id: I2899322622a4c3fd1d0ddc75b100b81669417660
|
|
* I had issues when folder name contained "armnn" and
a .tflite model was used, as the wrong parser was selected.
* Now only the extension, and not the full string, is
considered when selecting parser.
Change-Id: If7964d2ce5535f7d25762d2a2d7e810bf1a1ed43
|
|
* Some CL kernels are not run after the first inference and this breaks
the profiler which is expecting a measurement for every kernel each run
* Add a function HasKernelMeasurements() to ascertain if the Event is
returning kernel measurements and if so insert 0.0 values for any missing
kernel measurements.
* Fix ExecuteNetwork to only print a json object after all inferences
have completed
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I99f2bb0db847f5a52ab4c5705b072155c6b6f333
|
|
* Asserts are removed during a release build, which causes build
failures due to unused variables.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ie36f2bd22f5b2916b03ba7e64c1895fdf21f11f0
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I6c712d6b817f87f3d3a996e11e81208e9449da9f
|
|
* Update GEMM reshaped rhs only heuristic
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I3b7489c1133ed8ae03dbc7c1e30b1979d604f4ba
|
|
* As part of testing IVGCVSW-7122, pyarmnn tests were failing as py35
is no more the default version in Odroids. Also we do mention Python 3.6+
as the supported version in pyarmnn/README.md
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I39407ba8689fbb064c32841d16d66c5080acbc10
|
|
This reduces build steps to a single guide rather
than providing multiple options
- Remove Standalone build from documentation
- Remove Wheel installation from documentation
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: Ic1411eba5889434b2e1761ce2b9c8fdd2453b352
|
|
* Signed32 missing from CompareAndPrintOutput
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: If3c93fb0d73c566ddcf439fceaa6d629029df18f
|
|
* Fix for Debug mode in ExNet does not work with ConstTensorsAsInputs
* Remove unnecessary assertion with ambiguous message in LoadedNetwork
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I9cd5d1f811dbbc89072d1190c510bf1b22e3069c
|
|
* IRuntime.hpp includes one file from the profiling interface and it's
causing some build problems for some users. Replace the include with
a forward declaration of arm::pipe::ILocalPacketHandlerSharedPtr.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I6064e02db4ca056423bb448ddc0d712b50a704cb
|
|
* Fix overflow in NEActivationLayer for FP16 type
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I0f2e1609b8a46b69a9d83a4fa2cb6cf22af8c8d3
|
|
* This affects only to the layers (not workloads) Conv, DWConv and FC
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I66a91ed1a78bc0464e00423c7fc7c28c91d199ce
|
|
In ArmNNExecutor::ArmNNExecutor the call to m_Runtime->LoadNetwork was
ignoring the Status result and continuing to execute with a failed
network. In addition throwing an exception from the constructor resulted
in a segmentation fault.
* Modify IExecutor to allow the constructor to mark itself as failed.
* Modify ArmNNExecutor to mark itself as failed when LoadNetwork returns
an error.
* Modify ExecuteNetwork to check the value of m_constructionFailed.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: Idf222cb2b66e1051875dc67046734f2b00b288d1
|
|
* Add FP32 Neon™ swish activation
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I606855096d90aa16aef9282e5860d41e3dc64d7f
|
|
param contains a value of -1.
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: I538347083e9f22b3f3b6c048aebc2cf5cf4dc786
|
|
* Removing support for aarch32 builds from the
build-tool as Arm NN is dropping support for
this target architecture.
* Fixed missing additional ACL scons params in
ACL build.
Change-Id: Ib18b3286ce18d169ce8d8af2bd9b5d54f110eb3b
Signed-off-by: James Conroy <james.conroy@arm.com>
|
|
* Add test for ClGemmLowpMatrixMultiplyCore to test a batched matrix multiplication
with variable input tensors
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Id1531ff661de6a01c861737111924e4c23e814fb
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I1bc3d50b8fa6e216d8b6b7e3421d2ff37a21712c
|
|
* Creates delegate/src/BatchMatMul.hpp
* Add VisitBatchMatMul function
* Add BatchMatMul to switch in armnn_delegate
* Creates delegate/src/test/BatchMatMulTest.cpp
* Creates delegate/src/test/BatchMatMulTestHelper.hpp
* Add Int8 and Fp32 unit tests on ref backend
* Add BatchMatMul to delegate supported ops
Signed-off-by: Ryan OShea <ryan.oshea3@arm.com>
Change-Id: I50e61314cf063f986c8a0f7d508847a96953735e
|
|
* Update feature level support to FL6
* Add ConvertBatchMatMul function
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I93a77ba869bcddf432229a20e619304305d3982e
|
|
* Add a breakdown of the runtime options to the documentation.
* Exclude test classes from Doxygen generation.
* Limit the file extension selection for Doxygen generation.
* Add the support library to be generated.
* Fix some broken markups.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I81896f2d7fff503a1d51d6d4ac3876aa8b84118e
|
|
buffer data for Slice operator.
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: I86735faf8d57168a63ca6657f281ec03b90f7235
|
|
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I6f2b52f7277215c97be3d53969a9eaef137e3636
|
|
The subgraphs produced by SubgraphViewSelector were not produced in a
deterministic order, as the order was determined by the pointer values
of some objects, which are not guaranteed to be the same for each
execution.
This patch adds a post-processing sorting step based on the GUIDs of the
layers and the slot indices so that the results will be the same for
each execution.
This makes debugging the optimised graph much easier as subsequent
stages can also be deterministic. It also simplifies some unit tests.
Change-Id: I64f552706b7fb1bf82c19d85a448e054277917bc
Signed-off-by: Rob Hughes <robert.hughes@arm.com>
|
|
* Add test case for disable Winograd on fp16 if fast-math = false
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Id4543f566a095c5822aa0f7df5ad9af4c4bd739e
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I3f0637b4b2911c9cf314b3459ff84e959e467073
|