Gitiles
Code Review Sign In
review.mlplatform.org / ml / ComputeLibrary / 8a042116e26b702a74514b251b0961c4b661e6ca / src / runtime
  1. 368e635 COMPMID-1047 Extract Flatten function from Im2Col for NEON by Giorgio Arena · 6 years ago
  2. 6c95c2d COMPMID-1188: Static tuning of CLScale by Georgios Pinitas · 6 years ago
  3. 1049020 COMPMID-1188: Set all arguments to const in NEDepthwiseConvolutionLayer::validate() by Georgios Pinitas · 6 years ago
  4. 25f45a4 COMPMID-1060 LSTM FP32 NEON by Michalis Spyrou · 6 years ago
  5. 26b2216 COMPMID-1480 Add support for NHWC QASYMM8/FP32(non-optimized) to NEON DepthwiseConvolution by Giorgio Arena · 6 years ago
  6. 1d38259 COMPMID-1366 Implement NECopy by Michalis Spyrou · 6 years ago
  7. d30714a COMPMID-1188 - Fix CLWinogradConvolutionLayer for NHWC by Gian Marco Iodice · 6 years ago
  8. 916d1bc COMPMID-1498 - Enable grouping in CLGEMMConvolutionLayer by Gian Marco Iodice · 6 years ago
  9. 9fef38a COMPMID-1376: Add support for QASYMM8 in CLDeconvolutionLayer by Michele Di Giorgio · 6 years ago
  10. 72ccc44 COMPMID-1504: (Nightly) Segfaults on CL and android by Georgios Pinitas · 6 years ago
  11. e3f1bd1 COMPMID-1246 Remove unused window iterator from NERNNLayer. by Michalis Spyrou · 6 years ago
  12. db9d46d COMPMID-1485 - Add support for NHWC when running NEGEMMConvolutionLayer with FP16/QASYMM8 by Gian Marco Iodice · 6 years ago
  13. 0f17039 COMPMID-1342 Add grouping support to CLIm2ColKernel by Giorgio Arena · 6 years ago
  14. b6eb353 COMPMID-1478: Stop relying on static default OpenCL objects in cl2.hpp by Anthony Barbier · 6 years ago
  15. d8b03dd MLCE-13: Sanitizing matrix argument in the Warp. by Pablo Tello · 6 years ago
  16. dff601d COMPMID-1488 - Add support for NHWC when running CLGEMMConvolutionLayer with QASYMM8 by Gian Marco Iodice · 6 years ago
  17. 6b4e604 COMPMID-1303: CLDepthConvert : Add support for FP32 -> FP16 and FP16 -> FP32 + validate() function by Michele Di Giorgio · 6 years ago
  18. f2bd261 COMPMID-1188 - Removed the multiplication by 4 in NEGEMMInterleavedWrapper by Gian Marco Iodice · 6 years ago
  19. 215b4ea COMPMID-1277 - Optimizing CLIm2ColKernel for NHWC. by Gian Marco Iodice · 6 years ago
  20. 20394d5 COMPMID-1248 Enabled memory manager in NEWinogradConvolutionLayer by Anthony Barbier · 6 years ago
  21. 3d677cc COMPMID-1406: Refactor gemm_interleaved to use our own types and scheduler by Anthony Barbier · 6 years ago
  22. 597a856 COMPMID-872 - Rework NEGEMMConvolutionLayer to use NEGEMM by Gian Marco Iodice · 6 years ago
  23. 883f489 COMPMID-1475: (OCLGrind) FP exception in NEGEMMConvolution by Georgios Pinitas · 6 years ago
  24. 195b0ba MLCE-36: FC tranpose weights by Georgios Pinitas · 6 years ago
  25. ebbb7f2 COMPMID-1188 - Fixed CLGEMMConvolutionLayer/NEGEMMConvolutionLayer for NHWC by Gian Marco Iodice · 6 years ago
  26. 9036749 COMPMID-1445: Fixed NEON GEMM assembly dispatch for QASYMM8 by Anthony Barbier · 6 years ago
  27. 68a3f56 COMPMID-1276 - Allow GEMM to work with 3D input tensor by Gian Marco Iodice · 6 years ago
  28. ef91516 COMPMID-1359: (Nightly) CLCannyEdge failures by Michele Di Giorgio · 6 years ago
  29. ff4bbce COMPMID-1188 - Fixed get_convolution_method in CLConvolutionLayer by Gian Marco Iodice · 6 years ago
  30. 8bc745d COMPMID-1124: Validate CLLSTM by Georgios Pinitas · 6 years ago
  31. 40ed6d8 COMPMID-1440: Access original B in gemm assembly when not pretransposed. by Georgios Pinitas · 6 years ago
  32. eb84d6b COMPMID-1434: Fix NEWinograd for NHWC and sub-tensors by Georgios Pinitas · 6 years ago
  33. ef776a8 COMPMID-1386: Add FC convert weights on NEON by Georgios Pinitas · 6 years ago
  34. a855af1 COMPMID-1401 Implement NEFullyConnectedLayer for QASYMM8 by Giorgio Arena · 6 years ago
  35. 5a3ee4f COMPMID-1347: Simplified CPUInfo enum by Anthony Barbier · 6 years ago
  36. eaefd00 COMPMID-1419: Make NEGEMMAssemblyDispatch dynamically typed instead of templated by Anthony Barbier · 6 years ago
  37. c8e84b5 COMPMID-1405: Create our own gemm_native kernel / function. by Anthony Barbier · 6 years ago
  38. 3efb375 COMPMID-1347: Fixed fallback CPUInfo populate function and added latest CPUs by Anthony Barbier · 6 years ago
  39. 7d66a8e COMPMID-1386: Add support for converting weights for CL. by Georgios Pinitas · 6 years ago
  40. ae54e02 COMPMID-1364: Add support for NHWC in NEDepthConcatenateLayer by Georgios Pinitas · 6 years ago
  41. e155337 COMPMID-1357: Stop passing around raw pointers in NEWinogradConvolution by Anthony Barbier · 6 years ago
  42. 578225e COMPMID-1357: Port NEWinogradConvolution to use NEGEMMAssemblyDispatch by Anthony Barbier · 6 years ago
  43. e29acf1 COMPMID-1365: Add support for NHWC in CLDepthConcatenateLayer by Georgios Pinitas · 6 years ago
  44. 42447c1 COMPMID-1188: Fixes LSTM IO dimension requirements. by Georgios Pinitas · 6 years ago
  45. 67fd4e8 COMPMID-1381: Fix nightly build failure on armv7 by Anthony Barbier · 6 years ago
  46. 71d9b57 COMPMID-1381: Cleaned up the AssemblyHelper interface by Anthony Barbier · 6 years ago
  47. d1794eb COMPMID-1226 Extend CLMeanStdDev to support FP32 / FP16 by Michalis Spyrou · 6 years ago
  48. 7777b1a COMPMID-1388: Change default CLTuner to the one for the detected GPU by Georgios Pinitas · 6 years ago
  49. 7784c83 COMPMID-1167: Validation for NEDepthwiseConvolutionLayer by Abe Mbise · 6 years ago
  50. b03f7c5 COMPMID-1188 : Rename TNOX to Mali-G76 by Georgios Pinitas · 6 years ago
  51. cdb43bd COMPMID-1246: Fixed typo in MidgardTuner.h filename by Anthony Barbier · 6 years ago
  52. 42a3172 COMPMID-1124 : Fixes in CLLSTM layer by Georgios Pinitas · 6 years ago
  53. 876be2a COMPMID-1339 - Implementing Winograd Convolution Layer 1x5 and 5x1 kernels on OpenCL NCHW by Gian Marco Iodice · 6 years ago
  54. 09daf4d COMPMID-1246 Change CLLSTM in order to match android tests by Michalis Spyrou · 6 years ago
  55. 2fab0c2 COMPMID-1246 Fix CLDeconvolution arguments when calculating by Michalis Spyrou · 6 years ago
  56. 584252d COMPMID-1355: OpenMP compilation is broken. by Georgios Pinitas · 6 years ago
  57. 7485d5a COMPMID-970 : Remove QS8 / QS16 support by Vidhya Sudhan Loganathan · 6 years ago
  58. 17d6d3c COMPMID-1352: Disable support for 4D softmax layer. by Georgios Pinitas · 6 years ago
  59. d051e97 COMPMID-811 Add NHWC data format support for CL depthwise convolution by Giorgio Arena · 6 years ago
  60. f1c2bf0 COMPMID-1201 - Implementing Winograd Convolution Layer 1x3 and 3x1 kernels on OpenCL by Gian Marco Iodice · 6 years ago
  61. f3139dd COMPMID-1334 (Nightly) Fix std::bad_alloc error in 32-bit NEON runs by Giorgio Arena · 6 years ago
  62. d89940e COMPMID-1345: Switched from using mutexes to atomics in the CPP Scheduler by Anthony Barbier · 6 years ago
  63. 19ea419 COMPMID-809: Add NHWC data format on CLGEMMConvolutionLayer. by Georgios Pinitas · 6 years ago
  64. bb71fe5 COMPMID-1253: Nightly: Fix Canny Edge NEON failing by Michele Di Giorgio · 6 years ago
  65. 17b1230 COMPMID-1293: Handle aligned allocations by Georgios Pinitas · 6 years ago
  66. 7282d56 COMPMID-1287: Extending NEWinogradLayer test suite by Pablo Tello · 6 years ago
  67. 542e92d COMPMID-1067 NEON RNN FP32 / FP16 by Michalis Spyrou · 6 years ago
  68. 7221933 COMPMID-1145: (API) Introduce prepare() stage (NEON/CL/GLES) by Georgios Pinitas · 6 years ago
  69. 376c85f COMPMID-1180: Add support for bucket multi-threading (Part2) by Anthony Barbier · 6 years ago
  70. 8e74f44 COMPMID-911: Allow GEMM to work with 3D tensors by Isabella Gottardi · 6 years ago
  71. 0a88792 COMPMID-1222 Implementing CLArithmeticDivision - FP32 / FP16 by Michalis Spyrou · 6 years ago
  72. 7df2786 COMPMID-1162: Enable NHWC data layout support for NEWinogradConvolutionLayer - part1 by Pablo Tello · 6 years ago
  73. e083771 COMPMID-1160 Turn Graph hints into heuristics by Giorgio Arena · 6 years ago
  74. 5cb3773 COMPMID-1278: Make intermediate tensors resizable in CLSoftmaxLayer::validate() by Michele Di Giorgio · 6 years ago
  75. aaab602 COMPMID-1156: Allow memory reusage when trying to reallocate a smaller CLTensor by Michele Di Giorgio · 6 years ago
  76. 17812ba COMPMID-817: Tuner: Port kernels to new design. by Georgios Pinitas · 6 years ago
  77. 70ba7d6 COMPMID-1257: Allow retaining weights in CLDeconvolutionLayer by Michele Di Giorgio · 6 years ago
  78. b62280a COMPMID-1244: Allow retaining weights in CLGEMMConvolutionLayer and CLFullyConnectedLayer by Michele Di Giorgio · 6 years ago
  79. 513fe2e COMPMID-1251 - Update GEMM's heuristic selection on OpenCL by Gian Marco Iodice · 6 years ago
  80. b95e210 COMPMID-1209: Enable memory manager for the GEMM workspace buffer by Georgios Pinitas · 6 years ago
  81. 46da23f COMPMID-813 Add NHWC data format support for CL scale by Michalis Spyrou · 6 years ago
  82. 52ecb06 COMPMID-1180: Add support for bucket multi-threading (Part 1) by Anthony Barbier · 6 years ago
  83. df473ea COMPMID-1182: printf doesn't work by Georgios Pinitas · 6 years ago
  84. 1b99338 COMPMID-568: Implement Canny edge function for CL/NEON by Abe Mbise · 7 years ago
  85. 2345f43 COMPMID-1168: enabling memory manager for _gemm_output gives wrong results by Georgios Pinitas · 6 years ago
  86. 65f9982 COMPMID-1038: stoi and stoul cleanup. by Pablo Tello · 6 years ago
  87. 793f87d COMPMID-1176: Add FP16 support in CLDeconvolutionLayer. by Georgios Pinitas · 6 years ago
  88. fa7ad56 COMPMID-1163: NEON Scale NHWC failures by Georgios Pinitas · 6 years ago
  89. 96b86a9 COMPMID-959 - Fix Validate in Convolution Layer by Isabella Gottardi · 6 years ago
  90. bcedf51 COMPMID-993 Implement CL LSTM function by Michalis Spyrou · 6 years ago
  91. a3221e6 COMPMID-1106 Add fast math support in NEWinogradConvolutionLayer by Giorgio Arena · 6 years ago
  92. 20b4313 COMPMID-814: Add validate method to scale. by Georgios Pinitas · 6 years ago
  93. 4d1e8a2 COMPMID-1166: Fixed alpha/beta mixup for some merges. by David Mansell · 6 years ago
  94. 55b3d12 COMPMID-1137 OpenCL concatenate width by Michalis Spyrou · 6 years ago
  95. a8aef29 COMPMID-792 - Re-enabled Winograd on NEON in all graph examples. by Gian Marco Iodice · 6 years ago
  96. cb0010b COMPMID-1102 : Enable the use of 4x4 tile sizes in neon implementation of winograd conv. by Vidhya Sudhan Loganathan · 6 years ago
  97. a8a4d9e COMPMID-959: Fixed typo in CLPermute::validate() by Anthony Barbier · 6 years ago
  98. 73d4aef COMPMID-948: Add validation to NEL2NormalizeLayer by John Richardson · 6 years ago
  99. 393fa4c COMPMID-814: NEScale NHWC support by Georgios Pinitas · 6 years ago
  100. 3e3b371 COMPMID-959: bugfix: CLDeconvolutionLayer::validate fails if bias==NULL (#439) by wr0112358 · 6 years ago