Gitiles
Code Review
Sign In
review.mlplatform.org
/
ml
/
ComputeLibrary
/
2ef70cf813a20f12ad11900892c374edb9fa6ee4
« Previous
5fc07aa
COMPMID-2338: Remove CLGEMMInterleave4x4 and CLGEMMTranspose1xW
by Gian Marco Iodice
· 5 years ago
b9626ab
COMPMID-2243 ArgMinMaxLayer: support new datatypes
by Michalis Spyrou
· 5 years ago
0af4418
COMPMID-2282: Implement SIN operator for CL
by Michalis Spyrou
· 5 years ago
3db1ba9
COMPMID-2362: Add Quantization Layer to the graph API
by Isabella Gottardi
· 5 years ago
0eb16cf
Puts duplicate code into a function in NEBox3x3Kernel
by Alan Kelly
· 5 years ago
efcefea
Optimizes NEBox3x3FP16Kernel by removing unnecessary int to float conversions
by Alan Kelly
· 5 years ago
a400334
Optimizes NEBox3x3Kernel by replacing floating point computations with fixed point
by Alan Kelly
· 5 years ago
32bd4dd
IVGCVSW-2794: Return correct status for ReduceMean operations.
by Georgios Pinitas
· 5 years ago
ca82e62
Update function list
by Michalis Spyrou
· 5 years ago
c6608ac
COMPMID-2218: 19.05 Release notes
by Michalis Spyrou
· 5 years ago
b3204e7
COMPMID-2093: Implement CLGEMMNative
by giuros01
· 5 years ago
3113a27
COMPMID-2194: Add identity activation function to GLES
by Usama Arif
· 5 years ago
1097147
COMPMID-2335: Crash in scale_bilinear_nhwc_core
by Pablo Tello
· 5 years ago
6ac5992
COMPMID-2254 Implement NEAbsLayer
by Manuel Bottini
· 5 years ago
80e55db
COMPMID-2194: Add identity function to NEActivationLayerKernel.cpp
by Usama Arif
· 5 years ago
0681e3b
COMPMID-2041: Create GEMM helper file for OpenCL.
by Usama Arif
· 5 years ago
52c54f6
COMPMID-2270: Implement POW operator for CL
by Usama Arif
· 5 years ago
eb312ef
COMPMID-2267: Implement NEG for CL
by Usama Arif
· 5 years ago
f391fff
COMPMID-2302 NEDeconvolution: support for FP16
by Manuel Bottini
· 5 years ago
68c6a79
COMPMID-2331: Fix FFTConvolutionLayer when no bias is used.
by Georgios Pinitas
· 5 years ago
6a98a6e
COMPMID-2194: Refactor activation function macro in OpenCL. Change all activation calls to macro from activation_float_helpers.h
by Usama Arif
· 5 years ago
c613210
COMPMID-2333: (3RDPARTY_UPDATE) Update benchmarks
by Georgios Pinitas
· 5 years ago
deaed2d
COMPMID-2334: Extend search space for the EXHAUSTIVE OpenCL tuner
by Gian Marco Iodice
· 5 years ago
81e671e
COMPMID-2269: Implement POW operator for NEON
by Usama Arif
· 5 years ago
c255aa7
COMPMID-2263: Implement NELogLayer
by Usama Arif
· 5 years ago
4b5c588
COMPMID-2248 L2NormalizeLayer: negative axis
by Manuel Bottini
· 5 years ago
2388de1
COMPMID-2177: Fix clang-tidy warnings
by Michalis Spyrou
· 5 years ago
083a069
COMPMID-1995: Change CLTuner default mode to NORMAL
by Michalis Spyrou
· 5 years ago
613bf43
COMPMID-2326: (Nightly) NEDepthConcatenate bechmark failure
by Michalis Spyrou
· 5 years ago
7dbfc41
COMPMID-2177 Fix clang warnings
by Michalis Spyrou
· 5 years ago
b56d0cd
COMPMID-2325: (Nightly) CLDeconvolution mismatches
by giuros01
· 5 years ago
42c52a6
COMPMID-2082: (3RDPARTY_UPDATE) Update benchmarks
by Georgios Pinitas
· 5 years ago
e2428a0
COMPMID-2194: Refactor activation function macro in OpenCL. New Macros for activation.
by Usama Arif
· 5 years ago
f6e475c
COMPMID-2268: Implement NEG for NEON.
by Usama Arif
· 5 years ago
bcfd09a
COMPMID-2177 Fix clang warnings
by Michalis Spyrou
· 5 years ago
46a49a0
COMPMID-1635: Optimize CLDeconvolutionLayer - Part III
by giuros01
· 5 years ago
879e8dd
COMPMID-2115: Fix CLFuseBatchNormalization
by Manuel Bottini
· 5 years ago
154bc1c
COMPMID-1973: Implement FFTConvolutionLayer on NEON
by giuros01
· 5 years ago
ae1a89e
COMPMID-2118 : (Nightly) : CLGroupedGEMMConvolutionLayer validation issues
by Vidhya Sudhan Loganathan
· 5 years ago
976f11f
COMPMID-1995: Updated Doxygen. Included Warning that CLGEMMInterleave4x4 CLGEMMTranspose1xW will be deprecated in 19.08 release.
by Usama Arif
· 5 years ago
299fdd3
COMPMID-2177 Fix clang warnings
by Michalis Spyrou
· 5 years ago
dc5d343
COMPMID-2144 : Update documentation for the new OpenCL tuner
by Vidhya Sudhan Loganathan
· 5 years ago
a788c2f
COMPMID-2108: Fuse Activation Layer in CLDepthwiseConvolutionLayer3x3Kernels for F32
by Manuel Bottini
· 5 years ago
01bbacb
COMPMID-2176: Add dilation support in calculate_same_pad()
by Pablo Tello
· 5 years ago
05fb448
COMPMID-1963: Implement FFT (2D) on NEON
by giuros01
· 5 years ago
a4f378d
COMPMID-1995: Fix clang-tidy warnings
by Michalis Spyrou
· 5 years ago
8ec0bb6
COMPMID-2117 : Use FFT convolution if output feature map depth is less than input
by Vidhya Sudhan Loganathan
· 5 years ago
14c4e0f
COMPMID-1961: Implement FFT (1D) on NEON
by giuros01
· 5 years ago
098516b
COMPMID-1995: Fix memory management for weights transform in CLFFT
by Georgios Pinitas
· 5 years ago
cd78347
COMPMID-1995: Update clang-tidy rules
by Michalis Spyrou
· 5 years ago
050471e
COMPMID-1974 : Extend CLTuner to support different of level of tuning
by Vidhya Sudhan Loganathan
· 5 years ago
d038daf
COMPMID-2119: (Nightly) : CLDepthwiseConvolutuion asserts on validate
by Usama Arif
· 5 years ago
0a33a21
COMPMID-2114: CLBatchToSpace missing validation test
by Pablo Tello
· 5 years ago
ad5d7b0
COMPMID-2120: (Nightly) CLWinogradOutputTransform asserts on validate
by Georgios Pinitas
· 5 years ago
f790fdb
COMPMID-1995: Update documentation with kernel/function support
by Georgios Pinitas
· 5 years ago
881f2de
COMPMID-2048: Add support for dilation in NEDepthwiseConvolution.
by Usama Arif
· 5 years ago
557d4ae
COMPMID-1995: Fix Disabled suite in test framework.
by Georgios Pinitas
· 5 years ago
7179837
COMPMID-2112: Rework memory manager support in NEWinogradLayer
by Georgios Pinitas
· 5 years ago
e73686a
COMPMID-2047: Add support for dilation in CLDepthwiseConvolution.
by Usama Arif
· 5 years ago
6631ac2
COMPMID-2116: (Nightly) : CLWidthConcatenate fails on 32-bit for QASYMM8
by Georgios Pinitas
· 5 years ago
2899e00
COMPMID-2049: Add support for deconvolution for qasymm8 on NEON
by Usama Arif
· 5 years ago
dcd949d
COMPMID-1995: Fix release builds
by Georgios Pinitas
· 5 years ago
a9c4472
COMPMID-2051 Refactor shape_calculator::calculate_concatenate_shape
by Michalis Spyrou
· 5 years ago
2ec6c1e
COMPMID-2110: Enable CLGEMMLowpMatrixMultiplyReshapeOnlyRHSKernel in CLGEMMLowp
by Gian Marco Iodice
· 5 years ago
a851bba
COMPMID-1995: Improves DepthwiseConvolution for qasymm8.
by Georgios Pinitas
· 5 years ago
8f43d74
COMPMID-2063: New Winograd implementation
by Pablo Tello
· 5 years ago
9e4824c
COMPMID-2111: ConcatenateLayer API should accept an index instead of an enum
by Georgios Pinitas
· 5 years ago
d66094e
COMPMID-1995: Fix NEPoolingLayer for quantized 3x3
by Georgios Pinitas
· 5 years ago
17dae87
COMPMID-2113: Check correct extension for host memory import.
by Georgios Pinitas
· 5 years ago
2d0e521
COMPMID-1995: Fix Sconscript TypeError
by Michalis Spyrou
· 5 years ago
8be9148
COMPMID-1959: Implements 2D FFT on OpenCL
by Georgios Pinitas
· 5 years ago
a50e702
COMPMID-2012: Remove unnecessary templates from NEON kernels
by Michalis Spyrou
· 5 years ago
8d94269
COMPMID-1492 Create tests/validate_examples/graph_depthwise_convolution
by John Kesapides
· 6 years ago
165308c
COMPMID-1995: Fix examples SConsscript install_dir
by Michalis Spyrou
· 5 years ago
4d0351c
COMPMID-2057: Implement and test import memory interfaces.
by Georgios Pinitas
· 5 years ago
a4bba9c
COMPMID-1995: Fix 32-bit NEDepthwiseConvolution errors.
by Georgios Pinitas
· 5 years ago
3418ba5
COMPMID-1058: Implement DeepSpeech as a graph example
by Michele Di Giorgio
· 5 years ago
926afe1
COMPMID-2097: Implement a heuristic to dispatch CLGEMMReshapedOnlyRHS kernel from CLGEMM
by Gian Marco Iodice
· 5 years ago
8c57169
COMPMID-2062 Rework CL ConcatenateLayer
by Michalis Spyrou
· 5 years ago
da953f2
COMPMID-2098: Scope handling of memory group resources.
by Georgios Pinitas
· 5 years ago
4335a8c
COMPMID-2107 CLPoolingLayer build failure
by Michalis Spyrou
· 5 years ago
9f15c5e
COMPMID-2062 Rework NEON ConcatenateLayer
by Michalis Spyrou
· 5 years ago
7191aaa
COMPMID-2101: DenseNet issue on NHWC
by Georgios Pinitas
· 5 years ago
a52e4cf
COMPMID-2060: Support different qinfo in PoolingLayer
by Pablo Tello
· 5 years ago
4fbcac6
COMPMID-2102: Fix CLWinogradConvolutionLayer for 7x7
by Gian Marco Iodice
· 5 years ago
3e92459
COMPMID-2047: Reference implementation for dilation in DepthwiseConvolution.
by Usama
· 5 years ago
b54c644
COMPMID-2011: ResNet50 returns wrong result
by Georgios Pinitas
· 5 years ago
6afea90
COMPMID-1995: Add the whole content of include plus utils in install_dir
by Michele Di Giorgio
· 5 years ago
86cfffe
COMPMID-2099: Enable dummy threads in CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel
by Gian Marco Iodice
· 5 years ago
642680a
COMPMID-1983 : Optimize NEIm2Col with NHWC data layout
by Vidhya Sudhan Loganathan
· 5 years ago
3bfacb2
COMPMID-1318: Implementing Winograd 7x7 NHWC on OpenCL - Part III
by giuros01
· 5 years ago
dbfc2dc
COMPMID-2069: Rework CL ML layers to run exclusively on CL.
by Georgios Pinitas
· 5 years ago
881c684
COMPMID-1318: Implementing Winograd 7x7 NHWC on OpenCL -- Part II
by Michele Di Giorgio
· 5 years ago
cafec8f
COMPMID-1024 Investigate concatenation for RNN/LSTM OpenCL
by John Kesapides
· 6 years ago
108a95e
COMPMID-2067: Add batch size option in validate_cl_gemm example
by Georgios Pinitas
· 5 years ago
875a751
COMPMID-1995: Increase tolerance for winograd input transform tests
by Michele Di Giorgio
· 5 years ago
62251f7
COMPMID-2002: Implement CLGEMMLowpMatrixMultiplyReshapedOnlyRHS - Transposed
by Gian Marco Iodice
· 5 years ago
b4a44ff
COMPMID-1995: Removed layout checks from Reduction ops
by Pablo Tello
· 5 years ago
4a8ec80
Optimize CL DeconvolutionLayer-Part II: Add CLDirectDeconvolution function to be used by CLDeconvolution.
by giuros01
· 5 years ago
cadb368
COMPMID-1995: Fixed graph fusion mutator for float types.
by Georgios Pinitas
· 5 years ago
Next »