Age | Commit message (Collapse) | Author |
|
* Optimization that searches for layers with ConstantLayers as inputs.
* The layer member variables are then redirected to these ConstantLayers.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I24a2bf0e8575b808343e0bbe3897b344e94796ad
|
|
* This change is necessary because tflite uses a [1,H,W,I*M] format
and uses the I*M dimension for per axis quantization. Our previous
layout [M,I,H,W] can't handle the correlating quantization scales.
* Updates Onnx-, TfLiteParser and TfliteDelegate
* Updates the CpuRef, CpuAcc and GpuAcc backends
* Adjusts unit tests
* Adds test to ensure models with old layout can still be read and
executed
* Adds conversion function to previous layout [1,H,W,I*M] --> [M,I,H,W]
which can be used by backend developers
!android-nn-driver:5553
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ifef23368b8c3702cf315a5838d214f7dc13c0152
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I982ecd66ea3ed4d88934cd8254832eecb4a7adb4
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ib01629256309cfe17f341909d5b9bbbb09361422
|
|
* 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
|
|
* In AddBroadcastReshapeLayerImpl check if a constant layer has other
connections before modifying its output tensor shape.
* In ElementWiseBaseLayer replace an ARMNN_ASSERT with a proper error
message.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Id3f3796c260eede61f076660505257a8b65d93fc
|
|
Some models would add a PAD layer before a pooling when they can't
express their padding configuration as SAME or VALID.
ArmNN can potentially merge the two merge the two because pooling
layers are described with explicit padding.
The merge is possible if the extra padding is neutral in the combined
pooling operation. A merged operation can only fuse paddings in the
dimensions that accept explicit padding in a pooling operation, i.e. the
spatial dimensions.
Signed-off-by: Diego Lopez Recas <diego.lopez.recas@gmail.com>
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Icd54718dcd9e797c923456b7fa6e0213e288e668
|
|
* Change layer storage of ConstTensors to std::shared_ptr<ConstCpuTensorHandle>
* Change clone to share ConstTensor rather than copy
* Remove uses of non-const GetTensor() call
* Reduce scope of non-optimized network in ExeNet, so memory can be released after use
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Ibb2c7309d12411d21405bd6024c76bcdf5404545
|
|
This reverts commit 51ce7d487c761358de105f82ff90553570aedac0.
Reason for revert: https://jira.arm.com/browse/IVGCVSW-5798 LargeGraph_TENSOR_FLOAT32 CTS tests failures
Change-Id: Ib031a47f605340b2202ecf074ce96a8b54c51075
|
|
Some models would add a PAD layer before a pooling when they can't
express their padding configuration as SAME or VALID.
Arm NN can merge the two because pooling layers are described with
explicit padding.
Signed-off-by: Diego Lopez Recas <diego.lopez.recas@gmail.com>
Change-Id: Id048186db6a005e0257bfbc1406c3b0dab2cdd58
|
|
Optimisation when the input is a const tensor
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I8b1357bdefc45880d064d7e448af364ac8644c0d
|
|
* Add OptimizerOptions, NetworkProperties, DebugCallbackFunction
to DelegateOptions
* Enable OptimizerOptions when the network is being optimized
* Enable NetworkProperties when loading network
* Enable DebugCallbackFunction
* Add error message when loading network
* Log warning instead of error when operator is not supported but
could fallback to another backend
* Improve uint16_t CompareData
* Unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I353035afb442774bfeb1c62570a90755c2ceaf38
|
|
* Added FP16 DataType support to DetectionPostProcess
* For DetectionPostProcess layer output is always Float32 regardless of input type
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I21f63dd08f0863e9a98e105b3009bab3da1ab0c3
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Id0625c58dbeea79874bf986b70d136ed9390bf83
|
|
* FuseBatchNorm class has been added to facilitate testing
* Only Convolution2D FP32 being fused
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I049c4770946ddca21b08516d4c9f4d0d22bf9b45
|
|
* Remove AddBroadcastReshapeLayer from TfLiteParser
* Add AddBroadcastReshapeLayer as optimizer
* AddBroadcastReshapeLayer optimizer unit tests
* Load-scope dynamic tensor broadcasting unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I3549e85b71b41cbd4d96c0f1ece7887acbca76d1
|
|
Change-Id: Ia4b4bb3be0ed6e933c77d58f8e9879b1370e9537
Signed-off-by: Laurent Carlier <laurent.carlier@arm.com>
|
|
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I3563209dcb3db1b40cf2db3855adc631b5e323be
|
|
* 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
|
|
* 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
|
|
Convolution2d and FullyConnected layers
* Add InsertConvertFp32ToBf16LayersBefore
* Add ConvertWeight to ConvertFp32NetworkToBf16Impl for Conv2d and FullyConnected
* Allow different input and output when input is BF16 and output is FP32
Conv2d and FullyConnected layers
* Unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Ic8f92ff28edcae08a72a3114a28f50c4619f919b
|
|
* Add ReduceFp32ToBf16 to OptimizerOptions
* Add ConvertFp32NetworkToBf16
* Add utility functions to insert conversion layers
* Add constant conversion BF16 <-> FP32
* Unit tests
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Iaff77e20c721400b052cb37eb9ef6fe16d7abaff
|
|
!referencetests:229377
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ia9b360b4a057fe7bbce5b268092627c09a0dba82
|
|
* Changed some existing Permutation specific optimizations to also support Transpose
* Added MoveTransposeUp optimization
* Added TransposeAsReshape optimization
* Added tests for Transpose optimizations
* Added missing layer tests for Transpose
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I20d099b284861402ae94aaa5dbf34907327a485f
|
|
Change-Id: I2e0884c66855071eb3aa72b86de06c6ed6389d50
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
|
|
* Only apply the Optimization when the base ReshapeLayer is connected to
the child ReshapeLayer and no other Layer.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Iccd676d657f9e7c829813f1bec9c82db8745d069
|
|
* Moved the relevant armnnUtils headers to the new location:
include/armnnUtils
* Update the header usage throughout the source code
!android-nn-driver:2387
Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
Change-Id: I2ba15cebcacafad2b5a1a7b9c3312ffc585e09d6
|
|
Dequantize
* Check for output data type as well as input data type when determining
whether we should attempt to fall back to FP32 if FP16 is not supported
* Override output type for Dequantize in IsLayerSupported() instead of
input type
* Updated original input type from FP16 to FP32 in InsertConvertFp32ToFp16LayersAfter()
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com>
Change-Id: Ic6477fd17cea5a91bd8bf9ae0cf836520897d5b7
|
|
into DepthToSpace
This is only possible in some limited cases, but removes an extra
layer from the graph and so should improve performance in all cases.
Change-Id: I7b3e6ba5dacb4fdb816ad270edaecda1436ab4cf
Signed-off-by: Rob Hughes <robert.hughes@arm.com>
|
|
MoveAllConnections
This is called at a time when newReshape has nothing connected to
its output slot (as it has just been created) and so is a no-op.
The code comment indicated that the intention was to connect the newReshape
to its *input*, but that has already been done in the InsertNewLayer() call
above, so the comment was incorrect.
There is a unit test covering this case ("OptimizeConsecutiveReshapesTest")
Change-Id: I933d5d1c6eb32f5a8269fb5d7c809cd7c89680d1
Signed-off-by: Rob Hughes <robert.hughes@arm.com>
|
|
* The execution crashed because the weights of a convolution were
null during the network execution
* Copied the weights and bias when folding a pad layer into a
convolution
Change-Id: I3ae72143d04cac90d4f878cdf3b1a08b2f2a5c90
Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
|
|
* Added new optimization for folding pad layer into convolution2d layer following it
* Added new test in OptimizerTests.cpp
* Added new optimization into All optimizations
* Added call to new optimization in Optimize in Network.cpp
* Updated CMakeLists.txt
Signed-off-by: Nina Drozd <nina.drozd@arm.com>
Change-Id: I682e07c71bbd42c49c02dda30a848a9ab2b16e7e
|
|
* Modified optimizer to support debug mode via DebugLayer
Change-Id: Ic8f313778e55540c182cf99876c44a0823be04c6
|
|
Change-Id: I663a0a0fccb43ee960ec070121a59df9db0bb04e
|
|
* Added RefPermuteFloat16Workload to serve as a fallback when CL
does not support the required permute configuration for FP16
* Move Half.hpp to armnnUtils as the utils library should not be
including private header files from the armnn library
Change-Id: Ibf0f698451e8406f7ed7cce470dab60b6d16361d
|
|
* moving backends/ClWorkloads to backends/cl
* and moving pure Cl workload related code to
backends/cl/workloads
Change-Id: I019a3c6b4da5e7a23074bf03fb057e63199ad129
|
|
All changes are the same:
//
// Copyright © 2017 ARM Ltd. All rights reserved.
-// See LICENSE file in the project root for full license information.
+// SPDX-License-Identifier: MIT
//
Change-Id: I37eae011411133663ca9d2b059714d92f8bf8e24
|
|
|
|
|
|
Change-Id: Id3c11dc5ee94ef664374a988fcc6901e9a232fa6
|