Age | Commit message (Collapse) | Author |
|
This patch adds int32 and int64 ArgMax op support.
Current ARMNN already has ArgMax op but not used, and
it doesn't support int64 output type.
So this patch adds a new type, Signed64, and also adds
ArgMinMax computation function for int64 type support.
In default, output tensor type of ArgMax op is int64 in case of
tensorflow lite model so this patch makes a proper function - ArgMax op
for int64 or int32 - to be called according to parsed output_type value.
With this patch, ARMNN supports both types - int64 and int32 - for
ArgMinMax op.
Changelog v1:
- Check if output data type of ArgMinMax op is valid or not.
- Use template function to support int32 and int64 types of ArgMinMax function.
- Keep using Signed32 as default data type of m_Output_Type.
Change-Id: I7a8e7e38dd9e5acc81464571d8b4d51378fc7f14
Signed-off-by: Inki Dae <inki.dae@samsung.com>
|
|
* Replaced with armnn/utility/NumericCast.hpp
* Some exclusions in reference backend
* Excluded as requires float implementation in NumericCast.hpp
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I9e4e9cd502c865452128fa04415fd6f250baa855
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Icd4c84ca6250b6e6d616f64481c5ff60671be9c0
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Id72d2c2851adcc1dd8f00a6103642b16ebe3a964
|
|
* Update CL pin to f3ad9513dd46fca1d6c5e4550286480fdbaba056
IVGCVSW-5005: Fix Deprecated Functions in CL/NEON Resize Workload
* Add missing virtual function to NeonIntercepterScheduler
* Update CL/Neon Resize Workloads to use ScaleKernelInfo
* Update CL/Neon Resize workloads to set correct Sampling Policy for Half Pixels
IVGCVSW-4981: Change CL/NEON Softmax axis value
* Default value is now 0
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I8a14c6a64e5a24bc1a66a7d3685cd388c2245702
|
|
* Added Neon workload for Fill Operator
* Enabled Fill operator tests on Neon
* NEFill function does not have validate() function yet
IsLayerSupported() function return true at the moment
* Added INT32 supported type for CpuRef
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I17bf5ec13750f46322a30653e15ba2a514f61f08
|
|
Change-Id: Ia4b4bb3be0ed6e933c77d58f8e9879b1370e9537
Signed-off-by: Laurent Carlier <laurent.carlier@arm.com>
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I1a66fdad63cef16866d9dfcb8a339647f856e1d4
|
|
Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Iedfcf0ef487bd7836b1bc4ba8a0e4337dc4da391
|
|
* Change ComputeSoftmaxAclAxis to work with int and uint axis
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ibbfa9ec7e2f0416e6885673212a767419c871cca
|
|
* exchange boost::polymorphic_downcast with armnn::PolymorphicDowncast
* remove unnecessary includes of boost::polymorphic_downcast
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ie603fb82860fe05fee547dc78073230cc62b2e1f
|
|
* Add QASYMM8_SIGNED data type support to NeonTensorHandle
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Iae34f7d67de83642606ccd8c61a1b72df7f2bb3a
|
|
* Change boost assert to armnn assert
* Change include file to armnn assert
* Fix ARMNN_ASSERT_MSG issue with multiple conditions
* Change BOOST_ASSERT to BOOST_TEST where appropriate
* Remove unused include statements
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I5d0fa3a37b7c1c921216de68f0073aa34702c9ff
|
|
* Add NeonConvertBf16ToFp32Workload
* Add NeonConvertFp32ToBf16Workload
* Add BFloat16 type support to NeonConstantWorkload and NeonTensorHandle
* Add ConvertBf16ToFp32Weight when ConvertBf16ToFp32Layer is added
* Unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Id5b44a203add5e0c98c1ca4e2162115741b56644
|
|
* Deprecate ClGreaterWorkload
* Add ClComparisonWorkload to encompass all comparison operators
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ida0ed7f59899d75b0fe7de1e7433b1ade018c6f1
|
|
* adds convert from armnn HardSwish to Acl HardSwish
* adds EndToEnd tests for Ref, Cl and Neon
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: If997dad783ff45d7f061254e7e30bb69d4d4d564
|
|
* Added TransposeLayer
* Added CL, Neon and Ref Workloads
* Added Transpose utilities
* Added Serializer and Deserializer support
* Added Quantizer support
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I04c755ba7cb5b1edf72b3c9f3c0314878032e3c7
|
|
* Added CpuRef implementation
* Added Unit Tests
* Added Quantizer Test
* Enabled Tests for Neon and CL backends on fp32 only
* Added to Serializer
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Ic23e1797dbc9352b40678c389d7fe2b836b582ea
|
|
* Added Quantization Scheme for QAsymmS8
* Added Unit Tests for QAsymmS8
* Renamed QAsymm8 calls to QAsymmU8
Signed-off-by: Ryan OShea <Ryan.OShea2@arm.com>
Change-Id: I897b4e018ba1d808cc3f8c113f2be2dbad49c8db
|
|
!android-nn-driver:2671
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ifeb6be7812fbb98b37f2a1439bfd5a3215de2a62
|
|
Signed-off-by: Ryan OShea <Ryan.OShea2@arm.com>
Change-Id: I9652844a868ce8e05c0433c051e7079cf203c422
|
|
!android-nn-driver:2622
Change-Id: If99d3eff71ff66ba28af1e5af248299fe04511b9
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|
|
!android-nn-driver:2572
Change-Id: I8fe52ceb09987b3d05c539409510f535165455cc
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|
|
Change-Id: I649cd2304fb0040164763d31a12fc77c6c3bed87
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|
|
!android-nn-driver:2435
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I10ecd4a8937725953396805f33a3562a5384c4d4
|
|
* Create a public API for the common backend files
* Move OutputHandler to armnn internal
* Remove unused headers
Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I3e86d908b021e3561befa9d45158d87d2cbb18c0
|
|
to per-axis case
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: I66a8360b6d86e95325dee58927dcbe62ccf6ad58
|
|
* Add Guid to Workload
* Remove circular dependency
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Signed-off-by: janeil01 <jan.eilers@arm.com>
Change-Id: I15342fa7481c6bdc050e057dce2d74bba07fe2dd
|
|
* Fixed numerous CTS/VTS failures related to Quantization
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: If5c20256366e80b6b9bbc46b2a1c410a9b8c48e1
|
|
* Added support for QuantizedSymm8PerAxis to ArmComputeTensorUtils.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Ib8662f216bc4b6b54e0099780f73bcf6ef05384b
|
|
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: I0bb0e9da306eee3e19dc9967a6c8bb01da998deb
|
|
Signed-off-by: Jammy Zhou <jammy.zhou@gmail.com>
Change-Id: I12562c4a9671cd43884e7c0c023ca463a99c3aa2
|
|
* Enable fp_mixed_precision flag for the failing test case
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: If13552165eb6598a84d213b82847b56a8c5f2783
|
|
* Update ACL pin to include change of ArgMinMax NEON/CL output type to Signed32
!android-nn-driver:2013
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I2a0c80e0557f619a213790027efca8b84bf6e58a
|
|
The following NeuralNetworkTests tests were failing on GpuAcc and CpuAcc:
GeneratedTests.softmax_v1_2_relaxed
GeneratedTests.softmax_v1_2_quant8
GeneratedTests.softmax_v1_2_2
GeneratedTests.softmax_v1_2_relaxed_2
GeneratedTests.softmax_v1_2_quant8_2
The default value for Softmax axis parameter in Android is -1 but is 1 in ACL.
Detect and handle this in ArmComputeUtils.ComputeSoftmaxAclAxis.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Ibb0660e4cb0dc6bd4c804c4397fbd61f38acdd9c
|
|
* Increased MaxNumOfTensorDimensions and fixed issues related to its use
* Fixed issues caused by assuming 5d tensors are invalid
* Updated ArmComputeTensorUtils for 5d tensors
* Added 5d tensor unit tests for add, mul, stack and reshape (needed by IVGCVSW-3527)
Signed-off-by: Matthew Jackson <matthew.jackson@arm.com>
Change-Id: I5bcd64942d0d04efcc6c5acb240ad4b88e010743
|
|
* Pad value for QASYMM8 is no longer stored in quantized form.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I048e1d233353c0560ae03a7cc1ed5199295352bc
|
|
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: I7f4c722141837939fd8904c52e75704a15c8a5e3
|
|
Use it for intermediate tensors on reference backend.
Lays the groundwork for memory management in the reference backend.
Change-Id: I7d3ee132cac31bde70ae6e1b815f4f0b03d550a6
Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com>
|
|
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: I3b021c0828d30298d99ddb211c9aae17fe3636f0
|
|
* Added backend support for multiple types of TensorHandle factories
* Refactored the backend API to enable new tensor strategies
* Added mechanism to determine memory strategies during optimization
* Perform mem-copy only when Direct access is not found
* Explicitly deleted the copy-constructor from OutputSlot to prevent
accidental local copies that would cause the DisconnectAll to be
called by the destructor
Change-Id: I7e812c8e5e6c1c20db1c5932749ac70fd93db7f8
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
|
|
Change-Id: Ia68da09d8f0fb0a04af9cb61062d7edaa5f1b887
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|
|
* Add check if Sub-tensors cannot be used, call ACL function
* Add computation of SplitAxis from SplitterDescriptor
* Add NeonSplitterWorkload functions
* Modify IsSplitterSupported to call ACL validate function
if sub-tensor cannot be used
* Also check if quantization parameters match when using sub-tensors
* Add more unit tests for Splitter in TfParser and TfLiteParser
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I31e4c7d055117c83c65b598c4125442173242226
|
|
!android-nn-driver:968
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: I03ccb4842b060a9893567542bfcadc180bbc7311
|
|
* Compute Softmax Acl axis for Cl and Neon
* Add unittests for Softmax in 3D and 4D
* Correct input and output layer names in CaffeSqueezeNet inference test
Change-Id: I2d369d9a2db19c40f2af3341039dd33f0c5637b1
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
|
|
no longer work
* Added two master variables ARMNN_COMPUTE_CL_ENABLED and ARMNN_COMPUTE_NEON_ENABLED
to android-nn-driver/Android.mk to easily control backend support
* Setting either of those two new variables then properly sets the
corresponding build macro
* If a specific backend gets disabled, the corresponding source files
will be excluded from the build
* Unified the usage of the pre-compile macros
!android-nn-driver:613
Change-Id: I582ff73493b70ba9e22ca2e38d875a0f19566c8a
Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
|
|
Change-Id: I0ab3339e8803a3e4e700d8fec9883eccc524b31e
|
|
separate branch) and master
* Brings in all the changes done for the pluggable backends
* Added sub-graph support and tests
* Added precompiled layer support and tests
* Moved BackendSettings to a separate file
* Removed the backend-specific code
* Ported DebugLayer and associated functionality
* Included fixes to make those changes work with master
Change-Id: Id7028fa7917527b844628d5aff5732e3d94c0488
|
|
* Unified ArmNN's weight format to [ M, I, H, W ] for the depthwise convolution
* Added conversion utilities to permute/reshape the weights as appropriate
when using CL and Neon backends
* Updated the reference implementation of the convolution
* Updated the relevant unit tests accordingly
!android-nn-driver:459
Change-Id: I07d0818efa9d1ca1e5dad82983aac1fe78eadb18
|
|
Factor out new BuildArmComputeReductionCoordinates function
from CL backend into ArmComputeTensorUtils.
Update NEON LayerSupport and WorkloadFactory objects
Change-Id: Icc975ec699199bffafbdb207323df509d35e1e04
|