Gitiles
Code Review
Sign In
review.mlplatform.org
/
ml
/
ComputeLibrary
/
424eb5d0bf7f9fe890b06dc7109d570e31d59d68
/
tests
424eb5d
COMPMID-556: Fix dangling references in NEFillBorder
by Georgios Pinitas
· 7 years ago
e82cd72
COMPMID-556: Add SoftmaxLayer benchmarks
by Georgios Pinitas
· 7 years ago
4af62a0
APPBROWSER-289 DirectConvolution1*1 optimization for FP16, and fix a typo in PoolingLayer
by Xinghang Zhou
· 7 years ago
07c37f9
APPBROWSER-313: Performance improvement for softmax layer
by Joel Liang
· 7 years ago
5124be5
COMPMID-661: Convolution quantized (#32)
by Chunosov
· 7 years ago
9873ea3
COMPMID-556: Revert COMPMID-652
by Georgios Pinitas
· 7 years ago
b8ab997
COMPMID-617 Add validation window to CLSoftmaxLayer
by Giorgio Arena
· 7 years ago
b9d38ee
APPBROWSER-312 Fully connected performance optimization
by Frank Lei
· 7 years ago
397d58a
COMPMID-617: Add validate support for NEON ArithmeticLayer
by Ioan-Cristian Szabo
· 7 years ago
c7f9b89
COMPMID-722 - Support for vector-matrix in GEMMLowp (NEON)
by Gian Marco
· 7 years ago
923241e
APPBROWSER-314: Performance optimazation for BatchNormalizationLayer
by zhenglin
· 7 years ago
0a878ae
COMPMID-556: Added a rounding policy to the quantize function
by Jaroslaw Rzepecki
· 7 years ago
8795ffb
COMPMID-700: Adds CPPPermute function
by Georgios Pinitas
· 7 years ago
8108d2e
COMPMID-718: Added winograd benchmarking tests.
by Pablo Tello
· 7 years ago
b40879e
COMPMID-556: Preserve case-sensitivity of positional arguments
by Georgios Pinitas
· 7 years ago
5b73ce1
APPBROWSER-313: medium value and relative standard deviation
by Joel Liang
· 7 years ago
5a7e776
COMPMID-556: Rename CPP folder to reference
by Georgios Pinitas
· 7 years ago
2ecbada
IVGCVSW-656 : added support (and unit tests) for asymmetric padding in
by Jaroslaw Rzepecki
· 7 years ago
04a8f8c
COMPMID-692 Consistent names for the interfaces
by Giorgio Arena
· 7 years ago
58c5794
COMPMID-706 - Add GEMMLowp output stage for scaling by a fixed point number
by Gian Marco
· 7 years ago
754e952
COMPMID-617: Add validate support for NEON PixelWiseMultiplication
by Ioan-Cristian Szabo
· 7 years ago
8951933
COMPMID-687: Winograd layer.
by Pablo Tello
· 7 years ago
2995f5b
COMPMID-617 Add validation window to CLActivationLayer
by Giorgio Arena
· 7 years ago
0162436
COMPMID-684: 2D In-Map normalization support for CL
by Georgios Pinitas
· 7 years ago
45bcc3a
COMPMID-661: QASYMM8 support for fully connected layer.
by Georgios Pinitas
· 7 years ago
303be90
COMPMID-617: Add validate support for NEON BatchNormalizationLayer.
by Ioan-Cristian Szabo
· 7 years ago
236bfe7
COMPIMID-553: MobileNet use case.
by Georgios Pinitas
· 7 years ago
9c450cc
COMPMID-695: Update Phase and Validation Wrapping
by John Richardson
· 7 years ago
5420b28
COMPMID-710 - Fix CLTranspose and NETranspose when the input shape is a vector
by Gian Marco
· 7 years ago
6fdfaa8
COMPMID-713: Address failures in OCLGrind for CLDirectConvolution
by Georgios Pinitas
· 7 years ago
7062382
COMPMID-617 Add validation window to BatchNormalization, PixelwiseMultiplication, ArithmeticOps
by Giorgio Arena
· 7 years ago
d912fd8
COMPMID-617: Add validation to NEON functions.
by Georgios Pinitas
· 7 years ago
b6f182d3
COMPMID-556: Fix CLDepthwiseConvolution3x3 Kernel.
by Georgios Pinitas
· 7 years ago
2fdc409
COMPMID-592: Ported GEMMInterleave4x4 tests.
by Pablo Tello
· 7 years ago
c23b633
COMPMID-617 Add validation window to CLNormalizationLayer
by Giorgio Arena
· 7 years ago
9f26b3e
COMPMID-617 Add validation window to CLPoolingLayer
by Giorgio Arena
· 7 years ago
05288a2
COMPMID-697 - Rework GEMMLowp interface on OpenCL
by Gian Marco
· 7 years ago
e8895f8
COMPMID-694: merge opencl_timer flag with the opencl flag
by Anthony Barbier
· 7 years ago
53d8c5c
COMPMID-464 Implement Depthwise separable convolution on NEON
by Michalis Spyrou
· 7 years ago
f3dfa27
COMPMID-632 Assembly: Integrate gemmlowp assembly version
by Michalis Spyrou
· 7 years ago
f202e50
COMPMID-556 Improved indentation and error handling in format_doxygen.py
by Anthony Barbier
· 7 years ago
dbfb31c
COMPMID-556 Print tests list using printers rather than cout
by Anthony Barbier
· 7 years ago
b7b3153
COMPMID-464 Implement Depthwise separable convolution on NEON
by Michalis Spyrou
· 7 years ago
02bf80d
COMPMID-661: Fix scale border issue (#38)
by Daniil Efremov
· 7 years ago
6194145
COMPMID-694: Fix condition which decides whether or not to profile a run
by Anthony Barbier
· 7 years ago
b28f29d
COMPMID-652 - Graph examples shouldn't need to link against OpenCL
by Isabella Gottardi
· 7 years ago
a3b1b46
COMPMID-667: Add validation static method to NEON GEMMlowp
by Georgios Pinitas
· 7 years ago
fc35b51
COMPMID-580 - Implement reference and CL/NEON validation for Remap
by Sanghoon Lee
· 7 years ago
37908d9
COMPMID-560 - Validation mismatches Gaussian Pyramid Half Scale
by Gian Marco
· 7 years ago
93dcd83
COMPMID-645: MobileNet network and benchmark
by Georgios Pinitas
· 7 years ago
6b77e91
COMPMID-665 - NEON: Add QASYMM8 in place Activation layer
by Gian Marco
· 7 years ago
30902ed
COMPMID-617: Add validation methods to ML CL functions.
by Georgios Pinitas
· 7 years ago
77f0f87
COMPMID-621: Disable colours when printing to file
by Anthony Barbier
· 7 years ago
d7295b7
COMPMID-661: Add QASYMM8 support (and basic tests) to CLDepthwiseConvolution3x3 kernel (#28)
by Dmitry Savenko
· 7 years ago
540d008
COMPMID-556: Fixes bias in CLDirectConvolutionLayer to be int32.
by Georgios Pinitas
· 7 years ago
43ce898
COMPMID-670 - Extend Sobel tests
by Isabella Gottardi
· 7 years ago
c1294fa
COMPMID-494: Fixing values to const
by Sanghoon Lee
· 7 years ago
41caa62
COMPMID-683: Normalization layer API clarification.
by Georgios Pinitas
· 7 years ago
181e651
COMPMID-675: NEGEMMLowp Assembly Integration
by Pablo Tello
· 7 years ago
bc8fb06
COMPMID-556 Pretty-printing is enabled twice by default
by Anthony Barbier
· 7 years ago
7a49c79
COMPMID-661: issue# 23 Scale border fix (#26)
by Daniil Efremov
· 7 years ago
283c179
COMPMID-676: Rework TensorInfo building
by Georgios Pinitas
· 7 years ago
ce58a9f
COMPMID-622 Let the user choose the units for the instruments
by Giorgio Arena
· 7 years ago
82afedf
COMPMID-678 Align Convolution Interfaces
by Giorgio Arena
· 7 years ago
02aabcc
COMPMID-583 - Implemented reference implementation and validation tests (NEON and CL) for Histogram
by Isabella Gottardi
· 7 years ago
2f8e077
COMPMID-556: Don't measure performance for the first run
by Anthony Barbier
· 7 years ago
4c2dd54
COMPMID-671: Add global pooling layer support.
by Georgios Pinitas
· 7 years ago
fa33043
COMPMID-556: Fix double pretty printer registration.
by Georgios Pinitas
· 7 years ago
e75a02b
COMPMID-675 - Reworked NEGEMMLowp interface/function
by Gian Marco
· 7 years ago
87f21cd
COMPMID-556 Updated libnpy.hpp
by Anthony Barbier
· 7 years ago
5edbd1c
COMPMID-556: Add support to build arm64-v8.2-a for Android platform (clang compiler)
by Ioan-Cristian Szabo
· 7 years ago
84b51ad
COMPMID-666: Adds the ability to import backing memory in a Tensor(CPU).
by Georgios Pinitas
· 7 years ago
c5d5439
COMPMID-621 Add output options for json and pretty files
by Giorgio Arena
· 7 years ago
f450caa
COMPMID-661: softmax-uint8 implementation (#16)
by Chunosov
· 7 years ago
7068f99
COMPMID-631: Merge branches/gles_compute branch
by Anthony Barbier
· 7 years ago
d607375
COMPMID-663: Fixed JSON writer for raw_data
by Anthony Barbier
· 7 years ago
1c52992
COMPMID-562: Port Schaar to new validation
by John Richardson
· 7 years ago
7d6bf6a
COMPMID-650 Check we have all the DirectConvolution benchmarks
by Giorgio Arena
· 7 years ago
ea1c950
COMPMID-556: Allow building only specific validation tests using the option variable test_filter.
by Pablo Tello
· 7 years ago
4a4a164
COMPMID-556: When an EnumListOption contains some invalid values: still keep the valid ones
by Anthony Barbier
· 7 years ago
6e43349
COMPMID-663 Adding instrument to return timestamps of clGetEventProfilingInfo
by Anthony Barbier
· 7 years ago
9c45f56
COMPMID-556 - Added 5x5 Direct Convolution benchmark for AlexNet
by Gian Marco
· 7 years ago
ab14c15
COMPMID-663: Reworked / cleaned up instuments' measurements
by Anthony Barbier
· 7 years ago
33da05b
COMPMID-556 - Reduced number of batches to 2 for VGG-16 tests
by Gian Marco
· 7 years ago
764f39e
COMPMID-556 - Fix Warp Perspective validation
by Isabella Gottardi
· 7 years ago
eed841c
COMPMID-661: Add QAsymm8 for Reshape (#29)
by Daniil Efremov
· 7 years ago
af6204c
COMPMID-661: Add avgpool-uint8 support. Optimize avgpool-fp32 for Bifrost. (#13)
by Anton Lokhmotov
· 7 years ago
3a873a5
COMPMID-556 - Fixed tests for AlexNet
by Gian Marco
· 7 years ago
1a39f6d
COMPMID-669 - Add validation for different border modes
by Gian Marco
· 7 years ago
f9d3a0a
COMPMID-617: Add validation functions.
by Georgios Pinitas
· 7 years ago
d6afedc
COMPMID-661: softmax-fp32 optimisation (#14)
by Chunosov
· 7 years ago
6ff12a0
COMPMID-662: Integrated the new a64_s8_gemm_12x8 + dot product kernel into ACL.
by Pablo Tello
· 7 years ago
d621bca
COMPMID-661: directconv-uint8 (#20)
by Chunosov
· 7 years ago
5a99ddf
COMPMID-590: Add transpose_validation to the test framework.
by Ioan-Cristian Szabo
· 7 years ago
3faea25
COMPMID-617: Adds validation to CLPoolingLayer
by Georgios Pinitas
· 7 years ago
b5908c2
COMPMID-653 - Arithmetic Subtraction, add support different datatype
by Isabella Gottardi
· 7 years ago
ab18212
COMPMID-616 - Optimizing GEMMLowp on NEON intrinsics
by Gian Marco Iodice
· 7 years ago
adaae7e
COMPMID-647: Exclude padding pixels from averaging factor.
by Georgios Pinitas
· 7 years ago
7362f0d
COMPMID-464 Implement Depthwise convolution 3x3 on NEON
by Michalis Spyrou
· 7 years ago
3c5f949
COMPMID-575: Port Magnitude to new validation
by John Richardson
· 7 years ago
Next »