Gitiles
Code Review Sign In
review.mlplatform.org / ml / ComputeLibrary / 9e235628aff153937460599b3cfee292749cb5f7 / arm_compute
  1. 4d37b18 COMPMID-2819: Retain layout in NEIm2ColKernel by Georgios Pinitas · 4 years, 5 months ago
  2. 868282a COMPMID-2819: Retain layout during configuration for multiple functions. by Georgios Pinitas · 4 years, 6 months ago
  3. df97019 COMPMID-2819: Retain data layout in NESpaceToDepth and NEDepthToSpace by Georgios Pinitas · 4 years, 6 months ago
  4. c740d11 COMPMID-2819: Retain configuration step data layout to avoid side-effects. by Pablo Tello · 4 years, 6 months ago
  5. ad83b55 COMPMID-2819: Retain configuration step data layout to avoid side-effects. by Georgios Pinitas · 4 years, 7 months ago
  6. e6b3202 COMPMID-2855: CLReduceMean throws error for invalid configs by Pablo Tello · 4 years, 7 months ago
  7. 1df25a8 COMPMID-2855: NEReduceMean throws error for invalid configs by Pablo Tello · 4 years, 8 months ago
  8. 3e84bb6 COMPMID-2920: NEInstanceNormalization fails on NHWC validations by Manuel Bottini · 4 years, 8 months ago
  9. 951b8a4 COMPMID-2309 : CLConvolutionLayer: support QUANT8_SYMM_PER_CHANNEL filters by Vidhya Sudhan Loganathan · 4 years, 8 months ago
  10. cd4e9ab COMPMID-2452: Add mnist example by Isabella Gottardi · 4 years, 8 months ago
  11. ee8cf32 COMPMID-2905: Fixed benchmark graph segfault. by Pablo Tello · 4 years, 8 months ago
  12. 7a669a8 COMPMID-2921 Performance regression on NEON Int8 by Giorgio Arena · 4 years, 8 months ago
  13. 7b12bfb COMPMID-1730: Document best practices by Michele Di Giorgio · 4 years, 8 months ago
  14. 8217c8e COMPMID-2895: Remove QASYMM8_PER_CHANNEL data type by Georgios Pinitas · 4 years, 8 months ago
  15. 5dea19e COMPMID-2579: Fuse batch normalization with convolution and depthwise convolution at graph level on NEON by Gian Marco Iodice · 4 years, 8 months ago
  16. eaa01ab COMPMID-2671 use Signed32 for default output data type of ArgMinMax by Sang-Hoon Park · 4 years, 8 months ago
  17. 77d4252 COMPMID-2853: VGG16 regression for fp32 by Georgios Pinitas · 4 years, 8 months ago
  18. 5264b7d COMPMID-2576: Fuse activation in Winograd output transform. by Pablo Tello · 4 years, 8 months ago
  19. 68adf44 COMPMID-2893: Bare-metal build failures by Georgios Pinitas · 4 years, 8 months ago
  20. dbdea0d COMPMID-2308: NEConvolutionLayer: support QUANT8_SYMM_PER_CHANNEL filters by Georgios Pinitas · 4 years, 8 months ago
  21. 75d4733 COMPMID-2138: Create safe math integer functions by Michalis Spyrou · 4 years, 8 months ago
  22. 0fc2d7a COMPMID-2862: remove forward declaration of ReductionOperation by Sang-Hoon Park · 4 years, 8 months ago
  23. 7ae80a9 COMPMID-2549: Add GLES Runtime Context interfaces. by Georgios Pinitas · 4 years, 8 months ago
  24. 62eeb53 COMPMID-2266: [CL] add support for Log Softmax by Sang-Hoon Park · 4 years, 8 months ago
  25. 44bfc3f COMPMID-1671: Allow fp mixed precision in CLFCLayer. by Georgios Pinitas · 4 years, 8 months ago
  26. 2697fd8 COMPMID-2707: add keep_dims parameter to Reduction Operation by Sang-Hoon Park · 4 years, 9 months ago
  27. df4cf57 COMPMID-2306: CLDepthwiseConvolution: support for QUANT8_PER_CHANNEL_SYMM by Michele Di Giorgio · 4 years, 9 months ago
  28. 8b72199 COMPMID-1889: Fuse bias addition and output stage in CLFCLayer. by Georgios Pinitas · 4 years, 8 months ago
  29. d985378 COMPMID-2588: Optimize the output detection kernel required by MobileNet-SSD (~27% improvement) by Giuseppe Rossini · 4 years, 8 months ago
  30. 0726398 COMPMID-2501: Support multiplier > 1 during QASYMM8 requantization for Quantized LSTM by Manuel Bottini · 4 years, 8 months ago
  31. 05069f0 COMPMID-2515: Merge optimized depthwise convolution to the generic depthwise convolution function by Manuel Bottini · 4 years, 9 months ago
  32. 48b3ef8 COMPMID-2577: Fuse bias addition and activation in gemm assembly kernels by Georgios Pinitas · 4 years, 9 months ago
  33. d93e263 COMPMID-2708 NEDepthwiseConvolution Generic: support for QUANT8_PER_CHANNEL_SYMM by Giorgio Arena · 4 years, 9 months ago
  34. d24affe COMPMID-2265 add support for Log Softmax to NEON by Sang-Hoon Park · 4 years, 9 months ago
  35. 7c60c99 COMPMID-2486: Remove disabled compiler warnings by Michalis Spyrou · 4 years, 9 months ago
  36. db8485a COMPMID-2205: CL runtime context. by Pablo Tello · 4 years, 9 months ago
  37. a046e16 COMPMID-2600: Implement a new and generic depthwise convolution for CL QASYMM8 NHWC by Michele Di Giorgio · 4 years, 9 months ago
  38. 6bff195 COMPMID-2486: Remove disabled compiler warnings by Michalis Spyrou · 4 years, 9 months ago
  39. 0d54804 COMPMID-2572: Update the heuristic in CLGEMM for FP16 by Gian Marco Iodice · 4 years, 9 months ago
  40. b27e13a COMPMID-2685: [CL] Use Weights manager by Michalis Spyrou · 4 years, 9 months ago
  41. 58c71ef COMPMID-2257: Add support for QASYMM8 in NEGenerateProposals by Michele Di Giorgio · 4 years, 9 months ago
  42. d64a46c COMPMID-2699: Add support for QASYMM16 in NEQuantizationLayer by Michele Di Giorgio · 4 years, 9 months ago
  43. 79f88e6 COMPMID-2313: Implement CL INSTANCE_NORMALIZATION function by Manuel Bottini · 4 years, 9 months ago
  44. b94c368 COMPMID-2645: Detection Post-Process fixes by Georgios Pinitas · 4 years, 9 months ago
  45. 3d8fe49 INFPRF-609:Performance Issue of the Latest ArmCL by giuros01 · 4 years, 9 months ago
  46. 2aa7fd0 COMPMID-2601 [CL] add mixed precision support to PoolingLayer by Sang-Hoon Park · 4 years, 9 months ago
  47. 5c4a8e9 COMPMID-2592 Create a new kernel for CLPad with SYMMETRIC and REFLECT mode by Giorgio Arena · 4 years, 10 months ago
  48. 9637b2e COMPMID-2671: Change ArgMinMax NEON/CL output type to Signed32 by Michele Di Giorgio · 4 years, 9 months ago
  49. cecb0a7 COMPMID-2637 [CL] fix broadcast pixel-wise multiplication with 5D tensors by Sang-Hoon Park · 4 years, 9 months ago
  50. 879d131 COMPMID-2452: Fix compilation issues on armclang by Georgios Pinitas · 4 years, 9 months ago
  51. 0c17aa2 COMPMID-2571: Add mixed-precision support in CLGEMMReshaped for FP16 by Gian Marco Iodice · 4 years, 9 months ago
  52. 40958ad COMPMID-2663: Interface to unbound mappings. by Georgios Pinitas · 4 years, 9 months ago
  53. 6b612f5 COMPMID-2310: CLGenerateProposalsLayer: support for QASYMM8 by Michele Di Giorgio · 4 years, 10 months ago
  54. 05639f6 COMPMID-2571: Add support for FP16 in CLGEMMReshaped - part 1 by Gian Marco Iodice · 4 years, 9 months ago
  55. 1a569a3 COMPMID-2161 [NEON] Create IWeightManager class by Michalis Spyrou · 4 years, 10 months ago
  56. b785dd4 COMPMID-2670: [CL/GC] Create a test case for dynamic tensor support by Georgios Pinitas · 4 years, 9 months ago
  57. ee6454a COMPMID-2668: Add GEMM/GEMMLowp heuristic for Mali-G51 by Gian Marco Iodice · 4 years, 9 months ago
  58. 4c268b9 COMPMID-2681: Use NE/CL/Pad in NE/CL/GenerateProposalsLayer by Michele Di Giorgio · 4 years, 9 months ago
  59. c9564cb COMPMID-2257: Implement NEGenerateProposals. by Pablo Tello · 4 years, 10 months ago
  60. 769c638 COMPMID-2314: Implement NEON INSTANCE_NORMALIZATION function by Manuel Bottini · 4 years, 10 months ago
  61. 26014cf COMPMID-2649: Generalize MemoryGroup. by Georgios Pinitas · 4 years, 10 months ago
  62. caa7dee COMPMID-2641 [NEON] Create a test case for dynamic tensor support by Michalis Spyrou · 4 years, 10 months ago
  63. d87a7b2 COMPMID-2650: Add support for QASYMM16 in CLQuantizationLayer by Michele Di Giorgio · 4 years, 10 months ago
  64. e874ef9 COMPMID-2646: Wrap thread-related structure to ease bare-metal support by Georgios Pinitas · 4 years, 10 months ago
  65. c370c75 COMPMID-2570: Add hybrid precision flag in GEMMKernelInfo. by Georgios Pinitas · 4 years, 10 months ago
  66. 9285adb COMPMID-2599: Implement a new and generic depthwise convolution on OpenCL (Fp32/FP16-NHWC) by Gian Marco Iodice · 4 years, 10 months ago
  67. 70ad619 COMPMID-2635: Add support for QASYMM8 in CPPBoxWithNonMaximaSuppressionLimit by Michele Di Giorgio · 4 years, 10 months ago
  68. 9032ee3 MLCE-129: NEPad 30x slower than TensorFlow's implementation by Manuel Bottini · 4 years, 11 months ago
  69. ffd31de COMPMID-2246: NEBoundingBoxTransform support QASYMM16 by Pablo Tello · 4 years, 10 months ago
  70. 12833d0 COMPMID-2204: RuntimeContext interface for NEON functions. by Georgios Pinitas · 5 years ago
  71. ed0e35b COMPMID-2246: Implement NEBoundingBoxTransform. by Pablo Tello · 4 years, 10 months ago
  72. 4aff98f COMPMID-2247: Extend support of CLBoundingBoxTransform for QUANT16_ASYMM by Michele Di Giorgio · 4 years, 10 months ago
  73. 49be2e3 COMPMID-2644: Add is_dynamic flag to ITensorInfo. by Georgios Pinitas · 4 years, 10 months ago
  74. b19c55d COMPMID-2640: Fix performance regression for Resnet101 Int8 on NEON by Gian Marco Iodice · 4 years, 10 months ago
  75. 3f632f3 COMPMID-2418: CLDequantizationLayer support for QASYMM8_PER_CHANNEL by Michalis Spyrou · 4 years, 10 months ago
  76. ebe2e8c COMPMID-2318: Implement NEROIAlignLayer by Pablo Tello · 4 years, 10 months ago
  77. d75f9e9 COMPMID-2318: Implement NEROIAlignLayer by Pablo Tello · 4 years, 10 months ago
  78. 578a9fc COMPMID-2317: Implement CLROIAlignLayer by Michele Di Giorgio · 4 years, 10 months ago
  79. b9070a4 COMPMID-2605: Add asymmetric padding support for Deconvolution layer by Matthew Jackson · 4 years, 10 months ago
  80. 35ea9a7 COMPMID-2317: Add new QASYMM16 data type by Michele Di Giorgio · 4 years, 10 months ago
  81. 29a01c9 COMPMID-2417: NEDequantizationLayer support for QASYMM8_PER_CHANNEL by Michalis Spyrou · 4 years, 10 months ago
  82. fb0fdcd COMPMID-2460: Add ELU activation on NEON,CL,GLES by Georgios Pinitas · 4 years, 10 months ago
  83. c853021 COMPMID-2417: Add new QASYMM8_PER_CHANNEL data type by Michalis Spyrou · 4 years, 10 months ago
  84. 57a8961 COMPMID-2631: Deprecate CLGEMMMatrixAdditionKernel by Gian Marco Iodice · 4 years, 10 months ago
  85. 205eed8 COMPMID-2080 Create a new kernel for CLPad with CONSTANT mode by Giorgio Arena · 4 years, 11 months ago
  86. 70e890e COMPMID-2541 Remove DepthwiseSeparableConvolutionLayer for NEON/CL by Giorgio Arena · 5 years ago
  87. 670c04a COMPMID-2561: Fix GEMM heuristic by Gian Marco Iodice · 4 years, 11 months ago
  88. 1720358 COMPMID-2336 Extend validation for depthwise native and fix same pad calculator by Giorgio Arena · 5 years ago
  89. fed275d COMPMID-2546: Fix kernel name tag invalid ptr for NEGEMMAssemblyWrapperKernel.h by Gian Marco Iodice · 5 years ago
  90. a7acb3c COMPMID-1849: Implement CPPDetectionPostProcessLayer by Isabella Gottardi · 5 years ago
  91. d820db6 COMPMID-2545: Reduce tests required by GEMM (OpenCL) by Gian Marco Iodice · 5 years ago
  92. dfa5331 COMPMID-2544 Add deprecation note to DepthwiseSeparableConvolutionLayer for CL/NEON by Giorgio Arena · 5 years ago
  93. 607339f COMPMID-2527: Remove activation function validation in CLWinogradConvolutionLayer by Gian Marco Iodice · 5 years ago
  94. f3622be COMPMID-1979: Fuse Activation Function in CLGEMM - part 4 by Gian Marco Iodice · 5 years ago
  95. c4d5136 COMPMID-2493: Update qs8 in Depthwise assembly by Georgios Pinitas · 5 years ago
  96. 5d42b46 COMPMID-2467 Include IP information to the tuning config by Giorgio Arena · 5 years ago
  97. 883bad7 COMPMID-1849: Add DetectorPostProcess operator by Isabella Gottardi · 5 years ago
  98. bd9097d COMPMID-2336: Rename the new generic depthwise convolution on NEON by Gian Marco Iodice · 5 years ago
  99. 44f5572 COMPMID-2179 New generic depthwise convolution for NEON F32 NHWC by Giorgio Arena · 5 years ago
  100. cfa2bba COMPMID-2178: Update GEMM assembly code. by Georgios Pinitas · 5 years ago