Gitiles
Code Review
Sign In
review.mlplatform.org
/
ml
/
android-nn-driver
/
refs/tags/v21.11
/
ConversionUtils.hpp
tag
f520240921498114f452e33361b07dbff6e670dc
tagger
Nikhil Raj <nikhil.raj@arm.com>
Thu Nov 18 10:17:55 2021 +0000
object
bc8cacae736c569fd3caf343e7ad7a9e5534bf27
v21.11
1b46d13
IVGCVSW-6430 Clear up coverity issues
by Mike Kelly
· 2 years, 7 months ago
e2d611e
IVGCVSW-6428 Remove asserts
by Mike Kelly
· 2 years, 7 months ago
56c249c
BugFix: Fix skipping VTS tests.
by Matthew Sloyan
· 2 years, 10 months ago
29cc961
IVGCVSW-6158 ConstTensorsAsInput: FullyConnected - Tools
by Matthew Sloyan
· 2 years, 10 months ago
a20d2b8
IVGCVSW-5826 Change weights layout for depthwise to [1,H,W,I*M]
by Jan Eilers
· 3 years, 1 month ago
fc628d8
IVGCVSW-5410 Move ILayerSupport.hpp to backends folder
by Francis Murtagh
· 3 years, 1 month ago
a71c063
IVGCVSW-5826 Remove cross-wireing in depthwise
by Jan Eilers
· 3 years, 1 month ago
ccfeb5e
IVGCVSW-5766 Avoiding driver abort when Cts tests use invalid filter shape.
by Colm Donelan
· 3 years, 2 months ago
2e4a24a
IVGCVSW-5742 'NonConstWeights: Update FullyConnected in android-nn-driver'
by Sadik Armagan
· 3 years, 2 months ago
188675f
IVGCVSW-5665 Basic NN Driver support for next OS Version
by Sadik Armagan
· 3 years, 3 months ago
0182473
IVGCVSW-4874 Provide LayerSupportHandle to frontend users
by Francis Murtagh
· 3 years, 4 months ago
4e22f60
IVGCVSW-5267 Remove boost from core android-nn-driver
by James Ward
· 3 years, 7 months ago
7f492ac
IVGCVSW-5353 - Fix skipped concat VTS tests
by David Monahan
· 3 years, 7 months ago
4918446
IVGCVSW-5356 Fix skipping VTS DynamicOutputShape FullyConnected
by Finn Williams
· 3 years, 8 months ago
9b088d9
IVGCVSW-5304 Remove boost::numeric_cast from Android-nn-driver
by Matthew Sloyan
· 3 years, 8 months ago
fcf2a15
IVGCVSW-5245 Support Fused Activations for Dynamic Tensors
by Kevin May
· 3 years, 9 months ago
6e4081f
IVGCVSW-5270 Update ConvertConcatenation function to use ShapeInferenceMethod
by Keith Davis
· 3 years, 9 months ago
dbda4b7
IVGCVSW-5272 'Update ConvertLstm function to use ShapeInferenceMethod'
by Sadik Armagan
· 3 years, 9 months ago
346e811
IVGCVSW-5276 'Update ConvertSqueeze function to use ShapeInferenceMethod'
by Sadik Armagan
· 3 years, 9 months ago
291a16b
IVGCVSW-5220 Set all operations to unsupported after encountering an unsupported operation with with dynamic inputs
by Finn Williams
· 3 years, 9 months ago
81f27fd
IVGCVSW-5224 Fix for UnknownDimensionsTest in CpuRef and GpuAcc
by Kevin May
· 3 years, 9 months ago
9a04441
IVGCVSW-5220 Fix UnknownCombinationsTest/UnknownDimensionsTest
by Finn Williams
· 3 years, 9 months ago
4bd9a74
IVGCVSW-5182 Update Convert functions to use ShapeInferenceMethod. 1/2.
by Teresa Charlin
· 3 years, 9 months ago
a4983ce
IVGCVSW-4931 Update NN Driver to support dynamic tensors
by Finn Williams
· 3 years, 10 months ago
fc884b4
Add support for Rank and scalar tensors
by Finn Williams
· 4 years ago
00b586b
IVGCVSW-4920 Invalid Negative Stride fix
by Ryan OShea
· 3 years, 11 months ago
813f230
IVGCVSW-4453 Add Support for ANEURALNETWORKS_QLSTM to HAL 1.3 Driver
by Sadik Armagan
· 4 years ago
be6b3c2
IVGCVSW-4531 StridedSlice Failures on HAL Drivers
by Sadik Armagan
· 4 years ago
51e0b13
IVGCVSW-4675 Fix VTS 1.3 GeneratedTest.Test/armnn_resize*
by David Monahan
· 4 years, 1 month ago
7b9ce8d
IVGCVSW-4679 Fix VTS DeadlineTest failures 1.3 CpuAcc CpuRef
by Sadik Armagan
· 4 years, 1 month ago
185f588
IVGCVSW-4378 Fix transpose outputInfo for skipped Concat VTS in CL and Neon
by Teresa Charlin
· 4 years, 1 month ago
4d07e5e
IVGCVSW-4485 Remove Boost assert
by Narumol Prangnawarat
· 4 years, 2 months ago
51ba2c6
IVGCVSW-4407 HAL 1.3 Operators Support
by Sadik Armagan
· 4 years, 2 months ago
42477c1
IVGCVSW-4447 Add Hal 1_3 Support
by Kevin May
· 4 years, 2 months ago
793a70c
IVGCVSW-4565 TENSOR_BOOL8 data type not supported in AndroidNN Driver
by Sadik Armagan
· 4 years, 2 months ago
0b7a419
IVGCVSW-4482 Remove boost::ignore_unused
by Jan Eilers
· 4 years, 2 months ago
ec1e5b8
IVGCVSW-4473 Android R pre Hal 1_3 build changes
by Kevin May
· 4 years, 3 months ago
4a95658
IVGCVSW-4375 Add driver support for Transpose
by Mike Kelly
· 4 years, 3 months ago
eec836e
IVGCVSW-4375 Fixed driver crashes
by Mike Kelly
· 4 years, 3 months ago
26123db
MLCE-153 Padding calculation for TransposeConv ignores stride
by Mike Kelly
· 4 years, 4 months ago
6fd4ceb
IVGCVSW-4301 Correctly validate reshape for broadcastable inputs
by Derek Lamberti
· 4 years, 5 months ago
aed08ac
IVGCVSW-4262 Add Calls to IsReshapeSupported and IsPermuteSupported
by Kevin May
· 4 years, 5 months ago
1bde8e3
IVGCVSW-4260 Fix Transpose CpuRef VTS failures
by James Conroy
· 4 years, 4 months ago
00a963b
IVGCVSW-3813 Add Unary Elementwise Operation support to the android-nn-driver
by josh minor
· 4 years, 4 months ago
1a38cda
Rename quantized data types
by Derek Lamberti
· 4 years, 4 months ago
7b8d2e6
IVGCVSW-4315 Fix Fully Connected infer output shape bug
by FinnWilliamsArm
· 4 years, 4 months ago
7100649
IVGCVSW-4192 Add DEQUANTIZE to the android-nn-driver
by Keith Davis
· 4 years, 5 months ago
0182fd3
IVGCVSW-4225 Reject dynamic tensors to TRANSPOSE
by Matthew Bentham
· 4 years, 6 months ago
b9cb844
Catch std::exception instead of armnn::Exception
by Derek Lamberti
· 4 years, 6 months ago
00d6ed1
IVGCVSW-4209 Create a public API for the ArmNN Utils
by Matteo Martincigh
· 4 years, 6 months ago
c4aacb3
MLCE-115 Remove redundant support checking for Permute
by Matthew Bentham
· 4 years, 6 months ago
493e9b7
MLCE-144 Fix 2d pooling convert function
by FinnWilliamsArm
· 4 years, 6 months ago
aa5df2d
Refactor DequantizeIfRequired()
by Aron Virginas-Tar
· 4 years, 6 months ago
98c0f66
IVGCVSW-4151 HAL 1_2 Dequantize FP32 Per Channel Tests on CpuAcc Failing
by Sadik Armagan
· 4 years, 6 months ago
19fa0cc
IVGCVSW-3697 Add Support for ANEURALNETWORKS_ARG[MAX|MIN] to HAL 1.2 Driver
by Francis Murtagh
· 4 years, 6 months ago
d081194
IVGCVSW-4137 Failing LSTM android-nn-driver Unit Tests in HAL 1.2
by Sadik Armagan
· 4 years, 6 months ago
65a1b1d
IVGCVSW-4139 Fix regression in ConvertDequantize()
by Aron Virginas-Tar
· 4 years, 6 months ago
02dce09
IVGCVSW-4079 Modify the channel dimension accordingly with the weights permutation
by Teresa Charlin
· 4 years, 6 months ago
fb45e2f
MLCE-133 Fixed ASR hero use-case
by Pablo Tello
· 4 years, 7 months ago
9f0693b
IVGCVSW-3841 Add support for per-axis quantization
by Aron Virginas-Tar
· 4 years, 7 months ago
3b95960
MLCE-143 Fix for reference implementation for service stopping when running CTS.
by Teresa Charlin
· 4 years, 7 months ago
8f6429d
IVGCVSW-3818 Support Group convolution from Android
by Teresa Charlin
· 4 years, 8 months ago
8edb16d
IVGCVSW-3737 Add support for converting DEPTH_TO_SPACE
by Aron Virginas-Tar
· 4 years, 8 months ago
407718f
IVGCVSW-3746 Add Android NN Driver support for ABS
by Kevin May
· 4 years, 9 months ago
64b19b5
IVGCVSW-3618 Fix VTS/Neural Netwok Test Failures
by Sadik Armagan
· 4 years, 9 months ago
999e209
IVGCVSW-3643 Fix Ref Normalization failures
by Mike Kelly
· 4 years, 9 months ago
4627280
IVGCVSW-3633 Refactor HalPolicy to fully support V1.2 models
by Mike Kelly
· 4 years, 9 months ago
28e3d9f
IVGCVSW-3540 Fixed PReLU Being Skipped
by Mike Kelly
· 4 years, 10 months ago
c8bdb39
IVGCVSW-3604 Fix TransposeConv2d padding calculation
by Narumol Prangnawarat
· 4 years, 10 months ago
a0baa17
Fix explanatory comment around SanitizeBiasQuantizationScale()
by Aron Virginas-Tar
· 4 years, 10 months ago
b880520
IVGCVSW-3601 Fix skipped VTS Concatenate Tests
by Mike Kelly
· 4 years, 10 months ago
d74c505
IVGCVSW-3586 Fix Skipped Space_To_Batch Hal 1.2 VTS Failures
by Finn Williams
· 4 years, 10 months ago
0e4e439
IVGCVSW-3583 Correct data layout index for ConvertBatchToSpaceNd
by Finn Williams
· 4 years, 10 months ago
23b87b3
IVGCVSW-3583 Fix Skipped Batch_To_Space Hal 1.2 Tests
by Finn Williams
· 4 years, 10 months ago
0a87936
IVGCVSW-3591 Fixed unexpectedly skipped SUB tests
by Mike Kelly
· 4 years, 10 months ago
a5e2a45
IVGCVSW-3593 Remove HAL1.2-specific code from HAL1.0 ConvertConv2d() and ConvertDepthwiseConv2d()
by Aron Virginas-Tar
· 4 years, 10 months ago
e5604e6
IVGCVSW-3540 Fixed PReLU Being Skipped
by Mike Kelly
· 4 years, 10 months ago
15d63e2
IVGCVSW-3582 Fix Skipped Pooling Hal 1.2 Tests
by Sadik Armagan
· 4 years, 10 months ago
b7421e5
IVGCVSW-3578 Do not attempt to infer dynamic output shapes
by Aron Virginas-Tar
· 4 years, 10 months ago
000117b
IVGCVSW-3572 Fix tests failing due to usupported input tensor
by Aron Virginas-Tar
· 4 years, 10 months ago
6111316
IVGCVSW-3530 Fix DynamicOutput Tests for Android Q NeuralNetworks 1.0 & 1.1
by Sadik Armagan
· 4 years, 10 months ago
c921f6b
Fix verbose logging in ConvertPad
by Aron Virginas-Tar
· 4 years, 10 months ago
3c67394
IVGCVSW-3521 CpuAcc V1.2 pad Failures
by Mike Kelly
· 4 years, 10 months ago
2050c23
IVGCVSW-3530 Fix DynamicOutput Tests for Android Q NeuralNetworks 1.0 & 1.1
by Sadik Armagan
· 4 years, 10 months ago
573a8fa
IVGCVSW-3553 Fix failing zero_sized tests
by Aron Virginas-Tar
· 4 years, 10 months ago
d30093c
IVGCVSW-3477 Refactor android-nn-driver to use armnn ILayerSupported
by Ferran Balaguer
· 4 years, 11 months ago
86b36d4
IVGCVSW-3454 Fix VTS convolution test failures where output shape is incorrect
by Mike Kelly
· 4 years, 10 months ago
f03fcf0
IVGCVSW-3456 Add support for dynamic output shape in ConvertPrelu
by Aron Virginas-Tar
· 4 years, 11 months ago
cb8ac84
IVGCVSW-3301 Support PAD_V2 in HAL1.2 driver
by Aron Virginas-Tar
· 4 years, 11 months ago
0bd89a8
IVGCVSW-3370 Add broadcasting support to PReLU to properly run the Android VTS/NN
by Matteo Martincigh
· 4 years, 11 months ago
7a6d11b
IVGCVSW-3295 Support ResizeNearestNeighbour in HAL1.2 driver
by Aron Virginas-Tar
· 4 years, 11 months ago
44bcc02
IVGCVSW-3136 Run VTS tests with 1.2 Driver
by Sadik Armagan
· 5 years ago
cd700e4
IVGCVSW-3280 Refactor conversion methods to depend only on HalPolicy
by Aron Virginas-Tar
· 5 years ago
e205318
IVGCVSW-3265 Add support for optional Convolution2d dilation parameters
by Aron Virginas-Tar
· 5 years ago
07c7c9a
IVGCVSW-3272 Add support for optional DepthwiseConvolution2d dilation parameters
by Aron Virginas-Tar
· 5 years ago
b5fdf38
IVGCVSW-3181 Add HAL 1.2 support to android-nn-driver
by Mike Kelly
· 5 years ago
912b362
IVGCVSW-2911 Work towards compatibility with Q
by Matthew Bentham
· 5 years ago
0e7ab54
IVGCVSW-2947 Remove boost dependency from include/TypesUtils.hpp
by Aron Virginas-Tar
· 5 years ago
d5fd976
IVGCVSW-2886 Support multiple backends in Android driver
by Nattapat Chaimanowong
· 5 years ago
f29a2c5
MLCE-91 LSTM doesn't support optional input
by Kevin May
· 5 years ago
Next »