Age | Commit message (Collapse) | Author |
|
* Update Front-end and Tools.
* Updated Serializer, Deserializer and unit tests to reflect this.
* Updated TfLiteDelegate, TfLiteParser and OnnxParser.
* Updated Ref.
* Fixed resulting Neon / CL tests
* Unified optimizers for conv2d ops
* Optimizer Fix - Fp32ToBf16
* Partial implementation for ACL backends to fix VTS failures
!android-nn-driver:7477
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I5fb18877f7ee32643e15a9818945356274bb401b
|
|
!android-nn-driver:7418
* Update Front-end and Tools.
* Updated Serializer, Deserializer and unit tests to reflect this.
* Updated TfLiteDelegate, TfLiteParser and OnnxParser.
* Change NNDriver to new API.
* Updated Ref.
* Neon and Cl backend partially completed (Backend.cpp files).
* Added dynamic or constant input EndToEnd tests.
* Added ConstantTensorAsInputMemeberVariableRedirect Optimization.
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: Ib18b6c10a093042e165e25237dc04a4c67ba82da
|
|
Change-Id: Icc2f83c5f27f413758fee3e5c1445e9fc44f42c8
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
android-nn-driver do not execute.
* Change to src/backends/cl/workloads/ClLstmFloatWorkload.cpp fix LstmTests_GpuAcc tests.
* Change to src/backends/cl/workloads/ClConvertFp16ToFp32Workload.hpp & ClConvertFp32ToFp16Workload.hpp
fix MeanTests_GpuAcc and Convolution2DTests_1.1 tests.
* Added UnitTests to src/backends/cl/test/ClImportTensorHandleTests.cpp to test import on Convert Layers.
!android-nn-driver:7264
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: I0c46dc4b9c54eca8771ab12ed0302b6224606957
|
|
Change-Id: I18b8ca22896567904768170350ee5eb22edd4a22
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
Change-Id: I13353f50293eae565a75ccfda37209350512bbc6
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: Ib8e75eea49debe3b1dd8fa72623a55b26cb6ded4
|
|
Change-Id: I00521756c8a19d10bfdc98c6ef4204c7f84901c6
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
* Updated ABI version to 29 due to being the first ABI break in 22.05
!android-nn-driver:7226
Signed-off-by: Cathal Corbett <cathal.corbett@arm.com>
Change-Id: I9c50007dcd5b5e792757e7bd1213606df5ffec36
|
|
* Find and replace all workloads associated with imported IO
* Only attempt tensorhandle replacement if supported by all workloads
* Add new RefBaseWorkload to enable forced input for ref backend
* Store imported tensorhandles in preImportedTensorhandles instead of outputHandles
* Create pre-imported tensorhandles at network load-time
* Front load import workload validation to load network time
* Only call ReplaceTensorHandle when needed
Change-Id: I3816a71b7f57ae90388bb16462a75d4ef3544fa7
Signed-off-by: Finn Williams <finn.williams@arm.com>
|
|
* Update how we are setting the m_InputWorkloadSlotPairs to properly identify inputs
Signed-off-by: David Monahan <David.Monahan@arm.com>
Change-Id: I6d36f6d2bfdd42ade22c34e215f0d18b02702207
|
|
* Add AfterEnqueueWorkload to IBackendContext
* Implement AfterEnqueueWorkload in ClBackendContext to call Cl sync
* Set allocated data on outputhandler only once
* Handle PreImportedHandles and CurImportedId the same way as Async
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I9f59d57e298d4a494569faec3078d66af799f77b
|
|
* Add AllocatedData functions to OutputHandler
* Enable import aligned memory in ImportInputs
* Enable import aligned memory in ImportOutputs
* Allow to import input and output if the memory is aligned
* Implement Reconfigure function on ClConvolution2dWorkload
* End-to-end test on Ref and Cl to ensure that input and output memory
are imported when aligned
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I9e5e4c26d1ac2f1d806803ade5f64c6479c51718
|
|
* Added vectors to store the indexes of workload queues which contain inputs or outputs and their corresponding input / output slots
Signed-off-by: David Monahan <David.Monahan@arm.com>
Change-Id: Ide626726718216c7a778cce583da75af3ca2dc9a
|
|
Move the following header files from backendsCommon to armnn/backends.
* MemCopyWorkload.hpp
* TensorHandle.hpp
* Workload.hpp
* WorkloadData.hpp
* WorkloadFactory.hpp
Replace them with forwarding headers and a pragma deprecation message.
Resolve the deprecation messages in Arm NN code.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: I47f116b30f86e478c9057795bc518c391a8ae514
|
|
* enable external memory management for neon and ref backends
* change m_TensorMemoryVector to hold shared pointers
* change input layer backend Id to match backend id of connected layer
Signed-off-by: Finn Williams <finn.williams@arm.com>
Change-Id: I2216a724028312eb101b290df3f224177826b1a0
|
|
* Add parent LoadedNetwork profiling point
* Make generic populateParent function to print new descendents in json
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I41dc876bffae88e61a16d07fb13b062c321e78a6
|
|
Signed-off-by: Finn Williams <finwil01@e127804.cambridge.arm.com>
Change-Id: I84b52f8c7d7d0c1063c0daae0750dd820447aa6c
|
|
* This is to enable later work to instrument the Optimizer.
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
Change-Id: I2cf1fe022e0d100d6d8705adfbb8cab3ffc96a86
|
|
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
Change-Id: I2705fe04b5ec3a6046584d443f10d2d5cb3b5b87
|
|
* Add ClearImportedInputs/Outputs function to IRuntime
* Add UnImport function to ITensorHandle
* Remove mutex from IWorkingMemHandle
Change-Id: I34c9b6e1618755e10f3b4597afa1d9a9ea97e5fe
Signed-off-by: Finn Williams <finwil01@e127804.cambridge.arm.com>
|
|
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I257c3e79fa715170ac57caafc59131057c3a3b9c
|
|
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Ifc5e6f2e36767cb2a5cbf281d40ec9989b581abc
|
|
* Update acl and ref backend capabilities
* Add check for AsyncExecution capability in LoadedNetwork
* Add new exception for capabilities
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I2f053c845a64f336e96d452793639c6e0d3eaa89
|
|
* Add functionality to only output network details in ExNet
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I0c45e67193f308ce7b86f1bb1a918a266fefba2e
|
|
* Add switch for network details during profiling
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I8bd49fd58f0e0255598106e9ab36806ee78391d6
|
|
* Add new ProfilingDetails class to construct operator details string
* Add new macro which helps append layer details to ostream
* Add ProfilingEnabled to NetworkProperties so that profiling can be
realised when loading the network
* Add further optional info to WorkloadInfo specific to convolutions
* Generalise some JsonPrinter functions into JsonUtils for reusability
* Remove explicit enabling of profiling within InferenceModel as it is
done when loading network
* Add ProfilingDetails macros to ConvolutionWorkloads for validation
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: Ie84bc7dc667e72e6bcb635544f9ead7af1765690
|
|
!android-nn-driver:5802
* Extract the threadpool from LoadedNetwork/Runtime
* Refactor the threadpool to be handle multiple networks
* Trim IAsyncExecutionCallback and add an InferenceId to AsyncExecutionCallback
* Add AsyncCallbackManager class
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I36aa2ad29c16bc10ee0706adfeb6b27f60012afb
|
|
profiling/common
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I83de0fe9c83ec7e698817095ce5cfe3319c36ea6
|
|
allocations
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: If20572706d14df05e6d95ba436b364876d839495
|
|
* Allow the user to use create a tread pool with a single thread
* This is in keeping with how the android-nn-driver was implemented
* Add it to ExecuteNetwork thread pool creation
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I05b8048a9e0e45ae11d2b585080af28d9d008d81
|
|
* Remove concurrent flag from ExecuteNetwork as it is possible
to deduce if SimultaneousIterations > 1
* Add void RunAsync()
* Refactor some unit tests
Change-Id: I7021d4821b0e460470908294cbd9462850e8b361
Signed-off-by: Keith Davis <keith.davis@arm.com>
Signed-off-by: Kevin May <kevin.may@arm.com>
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I4e4eb107aa2bfa09625840d738001f33152e6792
|
|
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: Icc0d131c8ee2e9748e2f14762a75962b39c10f9d
|
|
* Generalises ConstCpuTensorHandle and inherited
classes by removing 'Cpu' from aliases.
* New renamed classes: ConstTensorHandle, TensorHandle,
ScopedTensorHandle, PassthroughTensorHandle,
ConstPassthroughTensorHandle.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I1824e0e134202735fb77051f20a7252f161dfe16
|
|
TensorHandleFactoryRegistry::GetFactory
* Modify Layer::CreateTensorHandles to include MemorySource
* Modify INetworkProperties to add MemorySource
* Disable Neon/Cl fallback tests until full import implementation complete
Change-Id: Ia4fff6ea3d4bf6afca33aae358125ccaec7f9a38
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
|
|
* Add inter layer memory management to WorkingMemHandle
* Change Const layers to be executed once in loadedNetworkConstruction
and share tensorHandle between all WorkingMemHandles
* Fix various reference workloads pointing to memory in the queueDescriptor
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I69d4b3c5c84d2f5abe4540c3e624ab4f00d88226
|
|
* Moved IAsyncNetwork into IRuntime.
* All LoadedNetworks can be executed Asynchronously.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Ibbc901ab9110dc2f881425b75489bccf9ad54169
|
|
!android-nn-driver:5042
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: Ia1ce8b839e81b46428ba0f78463e085e5906958d
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
|
|
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: If716f5f4e9b5433586b8a939d326830482da2f74
|
|
* Introduced CLCompileContext to ClWorkloadFactory
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ied38f4336210502e5f518b9955ae6a5ba3d242b3
|
|
CpuRef'
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I0c3ece5baf587e6cc22dfbec7ff98bd3573e0243
|
|
* Fallback to memory copy if memory import is not supported
* Remove direct compatibility between Neon and Cl Tensors
* Unit tests fallback from Neon to Cl and Cl to Neon
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Iec00a77423fb23b37a6b1aefee1b2ec4d649efca
|
|
* Replacing calls to boost:format with fmt:format.
* TensorUtils.cpp added outputShape.reserve call.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I4b2ed0f72039df824a2adca9309b8a9bbb158c5b
|
|
* If Output Layer is already connected to MemCopy Layer do not insert
CopyMemGenericWorkload.
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I9f813be5a3de2bc62d16864edb3eeaf371ef48e0
|
|
factory'
* Pass ModelOptions to WorkloadFactory
* Updated signature of CL and NEON Convolution2d workloads added FastMathEnabled param.
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I536178be8e4dd4083489e69febadaf0feeba46d2
|
|
* Allow memory import if padding is not required in Neon
* AddMockImportBackend for fallback tests
* Refactor GraphUtils
* Memory import unit tests
* Fallback unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Ic2e141e12774bf6d915e77745b6f6d2d83d9b82d
|
|
Change-Id: I6261f46404a3aab5c069bca40586994d31d26fe8
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
* Register TensorHandleFactories when create Workload Factory
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I6217c6f00d27b67e42b8043cbaf72d9db5789bfb
|
|
* Start inference profiling at the actual beginning
* Add profiling events for EnqueueInputs and EnqueueOutputs
* Add profiling event for working memory allocation
* Refactor Execute body to remove code duplication
* forward arguments to constructors rather than copy
Change-Id: Iacab85f0a02e88e2423885f86f97e4dba4037319
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|