Gitiles
Code Review
Sign In
review.mlplatform.org
/
ml
/
armnn
/
refs/tags/v20.08
/
src
/
backends
tag
81ffa0f5e9993b0454e7b9d64dce2174bdc71e53
tagger
Nikhil Raj <nikhil.raj@arm.com>
Fri Aug 28 14:15:55 2020 +0100
object
ba163f93c8a0e858c9fb1ea85e4ac34c966ef38a
v20.08
« Previous
586a9aa
IVGCVSW-4549 Add front end for new QLSTM layer
by James Conroy
· 4 years, 3 months ago
ea41b57
IVGCVSW-4582 Undefined reference to GetProfilingService causes failure in the Backends CI job
by Sadik Armagan
· 4 years, 3 months ago
b60dd24
IVGCVSW-4565 TENSOR_BOOL8 data type not supported in AndroidNN Driver
by Sadik Armagan
· 4 years, 4 months ago
ea54a01
IVGCVSW-4516 Add ConvertFp32ToBf16Layer and Ref workload support
by Narumol Prangnawarat
· 4 years, 4 months ago
a83af7b
IVGCVSW-4446 Add HardSwish EndToEnd tests
by Jan Eilers
· 4 years, 4 months ago
bca73e1
IVGCVSW-4444 Adding Elu end to end test
by Jan Eilers
· 4 years, 4 months ago
3184c90
IVGCVSW-4463 Change ProfilingService to a member of runtime from a singleton
by Sadik Armagan
· 4 years, 4 months ago
7ddbbae
IVGCVSW-4515 Add ConvertBf16ToFp32Layer and Ref workload support
by Narumol Prangnawarat
· 4 years, 4 months ago
0a99a14
IVGCVSW-4531 Fix for failing strided slice NNT/VTS tests on Android R
by David Monahan
· 4 years, 4 months ago
403a185
IVGCVSW-4512 Add BFloat16 Debug Workload
by Narumol Prangnawarat
· 4 years, 4 months ago
44179c3
IVGCVSW-4511 Add BFloat16 to RefLayerSupport and unit tests
by Narumol Prangnawarat
· 4 years, 4 months ago
ddb1d06
MLCE-159 Add QAsymmS8 to ArmnnQuantizer
by Francis Murtagh
· 4 years, 4 months ago
8eb2560
IVGCVSW-4482 Remove boost::ignore_unused
by Jan Eilers
· 4 years, 4 months ago
8832522
IVGCVSW-4517 Implement BFloat16 Encoder and Decoder
by Narumol Prangnawarat
· 4 years, 4 months ago
1ced464
MLCE-160 Error loading quantized model containing BatchNorm Layer
by Mike Kelly
· 4 years, 4 months ago
e32c844
IVGCVSW-4539 Add missing Activation Unit Test to CL and Neon
by Teresa Charlin
· 4 years, 4 months ago
9077473
IVGCVSW-4532 Update Acl Pin to fec13b8e20dc9d71d729f750633fa52c3a8a0036
by Kevin May
· 4 years, 4 months ago
3463eaa
IVGCVSW-4426 Add missing Neon tests for quantized types
by Keith Davis
· 4 years, 4 months ago
b015e5d
Modifying the IBackendProfilingContext interface to return a bool on
by David Monahan
· 4 years, 4 months ago
03fbeaf
IVGCVSW-4440 : Add HARD_SWISH Support to Activation in CpuRef
by Colm Donelan
· 4 years, 4 months ago
c9ea45a
IVGCVSW-4375 Add support for Transpose
by Mike Kelly
· 4 years, 4 months ago
3b3c381
IVGCVSW-4439: Adding Elu support to Activation
by David Monahan
· 4 years, 4 months ago
cf34f51
IVGCVSW-4368 Update Android.mk and backend.mk files with latest unit tests
by Colm Donelan
· 4 years, 4 months ago
d9fb6e2
IVGCVSW-4456 Disable NEON PostDetectionProcess
by James Conroy
· 4 years, 4 months ago
99bf80d
IVGCVSW-4456 Fix MobileNet SSD perf regression
by James Conroy
· 4 years, 4 months ago
67e6c54
IVGCVSW-4472 Yolo v3 ExecuteNetwork failing
by Keith Davis
· 4 years, 4 months ago
0c2eeac
IVGCVSW-4436 Add ExecuteNetwork test for mobilenet_v2_int8
by Keith Davis
· 4 years, 5 months ago
fcb802b
IVGCVSW-4340 Backend profiling: Add tests for Timeline functions
by Colm Donelan
· 4 years, 5 months ago
a856501
IVGCVSW-4420 Failing Unit Test Cl: Quantize layer
by Keith Davis
· 4 years, 5 months ago
032bc74
IVGCVSW-4338 Implement the Activation of Counters in backends
by Finn Williams
· 4 years, 5 months ago
c1536d6
IVGCVSW-4400 Backend Counter Registry Functionality
by David Monahan
· 4 years, 5 months ago
5c510f8
Fix legacy ClTuner API
by Derek Lamberti
· 4 years, 5 months ago
9b3e738
Clean up header inclusion in CL workloads.
by Matthew Bentham
· 4 years, 5 months ago
9add120
IVGCVSW-4386 Add ArmNN reference support for QAsymmS8
by Ryan OShea
· 4 years, 5 months ago
820eb14
MLCE-139: add capability to align corners for bilinear resize
by Sang-Hoon Park
· 4 years, 6 months ago
1aff393
IVGCVSW-4317 Implement the Profiling Context Initialisation
by Colm Donelan
· 4 years, 5 months ago
ef7e6b6
IVGCVSW-4389 Enable ClTuner in ClContextControl
by Derek Lamberti
· 4 years, 5 months ago
5e98b01
Clean up header usage a bit in NEON backend
by Matthew Bentham
· 4 years, 5 months ago
3d8bc97
IVGCVSW-4390 Refactor QUANTIZE to make use of Decoder/Encoder types
by Keith Davis
· 4 years, 5 months ago
867eba5
IVGCVSW-4399 Create Sample Dynamic backend
by Narumol Prangnawarat
· 4 years, 5 months ago
9d0ff74
IVGCVSW-4334 Wire Signed Quantized 8 bit type into ACL
by Keith Davis
· 4 years, 5 months ago
5e51cd8
IVGCVSW-4390 Refactor QUANTIZE to make use of Decoder/Encoder types
by Keith Davis
· 4 years, 5 months ago
c809a29
IVGCVSW-4388 Update ACL pin to 6a342648ae50beb8457871862f14fc9baef6b74f
by Teresa Charlin
· 4 years, 5 months ago
412424c
IVGCVSW-4149 Enable quantisation multiplier > 1 in all convolutions
by Ryan OShea
· 4 years, 5 months ago
e49755b
IVGCVSW-4316 First draft of IBackendProfiling and IBackendProfilingContext
by Colm Donelan
· 4 years, 5 months ago
1426a3f
IVGCVSW-4152 Remove check for QAsymm8 in ArgMinMax CL layer support
by Ryan OShea
· 4 years, 5 months ago
5204aa8
IVGCVSW-4306 Verify the tflite Yolo v3
by Keith Davis
· 4 years, 5 months ago
33d5827
IVGCVSW-4369 Add default offset and scale to IsLayerSupportedTestImpl
by Teresa Charlin
· 4 years, 5 months ago
5685802
IVGCVSW-4380 Add NEON backend support for BatchToSpaceNd
by Mike Kelly
· 4 years, 5 months ago
a023c40
IVGCVSW-4262 Use ACL Permute and Reshape Validate function in Neon and CL
by Kevin May
· 4 years, 7 months ago
e61f071
MLCE-156: Add Division workload in CpuAcc
by Pablo Tello
· 4 years, 5 months ago
0be3a88
IVGCVSW-4374 Add NEON backend support for SpaceToBatchNd
by Mike Kelly
· 4 years, 5 months ago
d466a54
IVGCVSW-4370 Deprecate DataType::QuantizedSymm8PerAxis
by Derek Lamberti
· 4 years, 5 months ago
4a3c610
IVGCVSW-4259 Add frontend and reference workload for UnaryOperationLayer
by josh minor
· 4 years, 6 months ago
246bd46
Remove inclusion of ArmNN.hpp where it is unnecessary.
by Matthew Bentham
· 4 years, 5 months ago
39ef3e5
Remove use of ArmNN.hpp where it is unnecessary.
by Matthew Bentham
· 4 years, 5 months ago
d2d917d
Add Signed32 to debug layer
by Narumol Prangnawarat
· 4 years, 6 months ago
f48afc6
Remove Logging.hpp from ArmNN.hpp
by Matthew Bentham
· 4 years, 6 months ago
f90c56d
Rename quantized data types to remove ambiguity for signed/unsigned payloads
by Derek Lamberti
· 4 years, 6 months ago
ba25aee
IVGCVSW-4246 Enable -Wextra by default
by Derek Lamberti
· 4 years, 7 months ago
baa177f
IVGCVSW-4246 Clean build of parsers with -Wextra
by Derek Lamberti
· 4 years, 7 months ago
901ea11
IVGCVSW-4246 Clean build of backends with -Wextra
by Derek Lamberti
· 4 years, 7 months ago
94a88d2
IVGCVSW-4246 Clean build Layers with -Wextra
by Derek Lamberti
· 4 years, 7 months ago
f2fa510
IVGCVSW-4214 Add README for standalone dynamic backend developer guide
by Narumol Prangnawarat
· 4 years, 7 months ago
e56fb45
IVGCVSW-4295 Disable mixed precision in ClPooling2dWorkload
by janeil01
· 4 years, 7 months ago
93cbbcc
IVGCVSW-4302 Depthwise CTS. Fix ReorderWeightChannelsForAcl.
by Teresa Charlin
· 4 years, 7 months ago
c374ff0
IVGCVSW-4246 Clean build of LayerTests with -Wextra
by Derek Lamberti
· 4 years, 7 months ago
df2a9b9
IVGCVSW-4261 Add backward compatibility to the Backend API
by Narumol Prangnawarat
· 4 years, 7 months ago
60a20fb
IVGCVSW-4212 Example of standalone dynamic reference backend
by Narumol Prangnawarat
· 4 years, 7 months ago
836b27b
IVGCVSW-4157 Pass custom options directly to backends
by Derek Lamberti
· 4 years, 8 months ago
fd27106
IVGCVSW-4211 Add Signed 8 bit Quantisation support into the Reference backend
by Finn Williams
· 4 years, 7 months ago
6a5e5e8
IVGCVSW-4227 Add CpuAcc backend support for DetectionPostProcess
by Derek Lamberti
· 4 years, 7 months ago
e5b8eb9
IVGCVSW-4210 Create a public API for the common backend files
by Matteo Martincigh
· 4 years, 7 months ago
9934e4e
Add asserts to check memcpy addresses
by Rob Hughes
· 4 years, 7 months ago
0844697
Replace boost logging with simple logger
by Derek Lamberti
· 4 years, 7 months ago
df31cfe
IVGCVSW-4070 Implement "send post-optimized network structure"
by Narumol Prangnawarat
· 4 years, 7 months ago
a68d853
IVGCVSW-3860 Per-channel QSymm8 support of DepthwiseConvolution2d in Cl
by Teresa Charlin
· 4 years, 7 months ago
e011d20
IVGCVSW-4209 Create a public API for the ArmNN Utils
by Matteo Martincigh
· 4 years, 7 months ago
2445744
IVGCVSW-3864 Add support for per-channel QSymm8 DepthwiseConvolution2d to Neon backend
by Aron Virginas-Tar
· 4 years, 7 months ago
9820d30
MLCE-115 Remove redundant error checking in CL backend
by Matthew Bentham
· 4 years, 7 months ago
710f664
IVGCVSW-4148 Extend reporting of quant multiplier > 1 as unsupported on ACL to per-axis case
by Aron Virginas-Tar
· 4 years, 7 months ago
899f64f
IVGCVSW-3859 Per-channel QSymm8 support of Convolution2d in Cl
by Keith Davis
· 4 years, 7 months ago
e9323ec
Add IsQuantizeType(DataType) for use in validation functions
by Aron Virginas-Tar
· 4 years, 7 months ago
94c4fef
IVGCVSW-3728 Add CL workload for Slice
by Aron Virginas-Tar
· 4 years, 7 months ago
21fc28b
IVGCVSW-3866 Add support for per-channel QSymm8 Convolution2d to Neon backend
by Aron Virginas-Tar
· 4 years, 7 months ago
e131a0b
IVGCVSW-4158 FP16 Mobilenet V1 and V2 30% regression on ArmNN on Mate20
by Sadik Armagan
· 4 years, 7 months ago
b86a382
IVGCVSW-4148 Report quant multiplier > 1 as unsupported for ACL
by James Conroy
· 4 years, 8 months ago
cf7c1c6
IVGCVSW-4151 HAL 1_2 Dequantize FP32 Per Channel Tests on CpuAcc Failing
by Sadik Armagan
· 4 years, 8 months ago
434de54
Revert "Only enable mixed precision FP16 pooling for Android Q"
by Kevin May
· 4 years, 8 months ago
52ec346
IVGCVSW-3697 Add check for ArgMinMax QAsymm8 to ClLayerSupport
by Francis Murtagh
· 4 years, 8 months ago
85ad78c
IVGCVSW-4068 Add Guid to Workload
by Narumol Prangnawarat
· 4 years, 8 months ago
036f02d
IVGCVSW-3729 Added neon slice workload and supporting neon layer tests
by josh minor
· 4 years, 8 months ago
49d0f12
MLCE-144 Cts NNAPI test cases failed
by Mike Kelly
· 4 years, 8 months ago
17d8d85
IVGCVSW-3486 Add clipping parameter validation in LstmQueueDescriptor
by janeil01
· 4 years, 8 months ago
87972be
IVGCVSW-4119 Fix FP16 to FP32 fallback mechanism in optimizer to work with Dequantize
by Aron Virginas-Tar
· 4 years, 8 months ago
5716de2
IVGCVSW-4140 Report per-axis quantization as unsupported for DepthwiseConvolution on ACL backends
by Aron Virginas-Tar
· 4 years, 8 months ago
30db8ad
NNXSW-1853 Change SubgraphViewSelector algorithm
by Rob Hughes
· 4 years, 8 months ago
721b82f
Print CMake messages on stdout rather than stderr
by Rob Hughes
· 4 years, 8 months ago
db1a283
Add FP16 support to DebugWorkload
by Aron Virginas-Tar
· 4 years, 8 months ago
60538ad
Only enable mixed precision FP16 pooling for Android Q
by Derek Lamberti
· 4 years, 8 months ago
Next »