Gitiles
Code Review Sign In
review.mlplatform.org / ml / ethos-u / ethos-u-vela / HEAD / ethosu / vela / tflite_graph_optimiser.py
  1. be78a05 MLBEDSW-8561: Striding support in H/W for StridedSlice by Rickard Bolin · 4 months ago main 3.12.0.rc1
  2. f493700 MLBEDSW-8969: Enable weight buffering for fully connected with batch shape by Johan Alfven · 4 weeks ago
  3. 7647b0f MLBEDSW-8875: MLCE: Update criteria when to move SplitSpliceRead to consumer by Johan Alfven · 7 weeks ago
  4. 9341bf4 MLBEDSW-8749: MLCE: Output diff on strided slice by Johan Alfven · 3 months ago
  5. fdbb072 MLBEDSW-8491: Add support for Mirror pad by Rickard Bolin · 9 months ago
  6. a8fda88 MLBEDSW-8290: MLCE: Add TRANSPOSE support by Johan Alfven · 7 months ago 3.10.0.rc1
  7. 67daf2a MLBEDSW-8219: Activation can not be fused with dma operation by Johan Alfven · 7 months ago
  8. afb56ae MLBEDSW-8201: [MLCE] Extended stride support for CONV_2D by Johan Alfven · 7 months ago
  9. 906c9e8 MLBEDSW-8042: MLCE: Add SQUARED_DIFFERENCE support by Johan Alfven · 12 months ago
  10. b4e804b MLBEDSW-8010: Refine fixup_pool_strides to also check stride by Johan Gunnarsson · 9 months ago
  11. 7ccc583 MLBEDSW-8003: Limit fixup_pool_strides to AvgPool and MaxPool by Johan Gunnarsson · 9 months ago
  12. c0bb868 MLBEDSW-7997: [MLCE] Extended stride support for TRANSPOSE CONV by Johan Alfven · 9 months ago
  13. 24570f0 MLBEDSW-7968: Add fixup for strides when kernel size equals IFM shape by Johan Gunnarsson · 9 months ago
  14. 9855637 MLBEDSW-7881: Convert Quantize op to Avgpool op in graph optimiser by Johan Gunnarsson · 9 months ago
  15. a71efe0 MLBEDSW-7754: Performance estimator is not using write/read shapes by William Isaksson · 10 months ago 3.9.0.rc1
  16. 9cf63a3 MLBEDSW-7756: MLCE: Grouped convolutions runtime problem by Tim Hall · 11 months ago
  17. da8741a MLBEDSW-7653: Extend Mean support for depth axis by Alexander Hansson · 11 months ago
  18. 1d5e859 MLBEDSW-7652: Add mean support for batch and channel when shape is 1 by Alexander Hansson · 11 months ago
  19. 6620714 MLBEDSW-7449: Add function description and type annotations by Raul Farkas · 12 months ago
  20. 90c34b5 MLBEDSW-7654: Extend support for Mean where HxW > 4096 by Alexander Hansson · 12 months ago
  21. 3e7157b MLBEDSW-7315: Add support for AvgPool with stride_width > 3 by Raul Farkas · 1 year ago
  22. 3b64f06 MLBEDSW-7648: Fix bug with filter padding in conv2d by Raul Farkas · 1 year ago
  23. 8e525ca MLBEDSW-7748: Add RSQRT support by Johan Alfven · 1 year ago
  24. 5ff4cd1 MLBEDSW-7223: Fusing Pad and AvgPool causes diff by Tim Hall · 1 year ago
  25. 10d6b3b MLBEDSW-7283: Add opt cases for strided CONV2D by Raul Farkas · 1 year, 4 months ago
  26. 69782af Revert "MLBEDSW-6343: Remove op_index constraint" by Raul Farkas · 1 year ago
  27. ce50273 MLBEDSW-2082: Add Exp support by Johan Alfven · 1 year, 1 month ago
  28. a68b82a MLBEDSW-7527: Mean operator output diff by Rickard Bolin · 1 year, 1 month ago
  29. 2180a17 MLBEDSW-7408: MLCE: Crash when serialising model LSTM by Tim Hall · 1 year, 2 months ago
  30. 7b3008a MLBEDSW-7487: Updated implementation for the Mean op by Johan Alfven · 1 year, 1 month ago
  31. 0ac0804 MLBEDSW-7196 Add LSTM support by Fredrik Svedberg · 1 year, 1 month ago
  32. c1ad80b MLBEDSW-7437: Add 64-bit output support for ArgMax by Johan Alfven · 1 year, 2 months ago
  33. 347c57b MLBEDSW-7442: Removed ofm quantization for ArgMax by Johan Alfven · 1 year, 2 months ago
  34. 56811e6 MLBEDSW-7439: Add support for input dims < 4 for ArgMax by Johan Alfven · 1 year, 2 months ago
  35. 72c6a24 MLBEDSW-6343: Remove op_index constraint by Raul Farkas · 1 year, 2 months ago
  36. 6986a07 MLBEDSW-6435: Implement support for ArgMax along depth dimension by Rickard Bolin · 1 year, 5 months ago
  37. a5e1b62 MLBEDSW-7312: Refactoring bypass_memory_only_ops by Johan Alfven · 1 year, 4 months ago
  38. 46c9477 MLBEDSW-7211: Convert fixup_asymmetric_weights to supported ops check by wilisa01 · 1 year, 3 months ago
  39. 16b5e5e MLBEDSW-7342: Regression: Output diff for mlperf_deeplabv3_mnv2_ade20k_int8 by wilisa01 · 1 year, 3 months ago
  40. 8289d51 MLBEDSW-4960: convert_resizebilinear_1x1_to_add creates constant output tensor by wilisa01 · 1 year, 4 months ago
  41. 090f18a MLBEDSW-7237: CONV_2D stride 4 optimisation by Raul Farkas · 1 year, 4 months ago
  42. 3b1578e MLBEDSW-7151: MLCE: Difference in model output between x86 & aarch64 by Tim Hall · 1 year, 4 months ago
  43. 5fdcf17 MLBEDSW-7203: Data type alias deprecations by Rickard Bolin · 1 year, 5 months ago
  44. 79a8904 MLBEDSW-6915: MLCE - Missing operators in Debug DB by wilisa01 · 1 year, 7 months ago 3.6.0.rc2
  45. bc6ee58 MLBEDSW-6620: Update copyright notice and years by Rickard Bolin · 1 year, 7 months ago
  46. ea4ba66 MLBEDSW-6905: Add dilation greater than 2 support by Tim Hall · 1 year, 6 months ago
  47. f3c7d55 MLBEDSW-6881 SHAPE single op network is optimised to nothing by Fredrik Svedberg · 1 year, 7 months ago 3.6.0.rc1
  48. 9d51ec4 MLBEDSW-7074: Updated reference kernel for the MEAN op by Johan Alfvén · 1 year, 7 months ago
  49. 65835e0 MLBEDSW-6941: Set correct OFM shape for fc op by Johan Alfvén · 1 year, 7 months ago
  50. a64616c MLBEDSW-6794: ResizeNearestNeighbor with HPC by Johan Alfvén · 1 year, 7 months ago
  51. 4a434cb MLBEDSW-6969 Remove RescaleAdd and RescaleMul operators by Fredrik Svedberg · 1 year, 8 months ago
  52. 1e5456f MLBEDSW-6708 Check the bias tensor in graph optimiser mean op by Fredrik Svedberg · 1 year, 8 months ago
  53. e84ed6b MLBEDSW-6962: MEAN height is greater than max kernel height by Johan Alfvén · 1 year, 8 months ago
  54. 017b4cc MLBEDSW-6928: Add int16 support for Resize Bilinear HPC by Rickard Bolin · 1 year, 8 months ago
  55. fea1516 MLBEDSW-6686: Resize bilinear HPC with tile padding by Rickard Bolin · 1 year, 11 months ago
  56. cc219be MLBEDSW-4338 Randomized int16 PAD output diff by Fredrik Svedberg · 1 year, 8 months ago
  57. 3642431 MLBEDSW-6938 Fix PReLU optimisation by Fredrik Svedberg · 1 year, 8 months ago
  58. 7f3ccd5 MLBEDSW-6929 Fix LeakyReLU int16 regressions by Fredrik Svedberg · 1 year, 8 months ago
  59. 701ba91 MLBEDSW-6869 Improve LeakyRelu support by Fredrik Svedberg · 1 year, 8 months ago
  60. 9ae3455 MLBEDSW-6613: Implement tile padding by Rickard Bolin · 2 years ago
  61. 0591663 MLBEDSW-6909: Use int32 acc for the Mean op by Johan Alfvén · 1 year, 9 months ago
  62. 6659165 MLBEDSW-6870 Optimisations for PReLU by Fredrik Svedberg · 1 year, 9 months ago
  63. 1700939 MLBEDSW-5029: Output diff for Mean op by Johan Alfvén · 1 year, 9 months ago
  64. 8ddd489 MLBEDSW-6832 PReLU support in Vela by Fredrik Svedberg · 1 year, 9 months ago
  65. 885033b MLBEDSW-4157: Add RESIZE_NEAREST_NEIGHBOR support by Tim Hall · 1 year, 10 months ago
  66. 47c7636 MLBEDSW-6616: ResizeBilinear align corners is incorrect by Tim Hall · 1 year, 10 months ago
  67. 1156317 MLBEDSW-6703 Add SHAPE operator to supported operators by Fredrik Svedberg · 1 year, 11 months ago
  68. a04f2f7 MLBEDSW-6687 Vela crashes in npu_serialisation.py and tflite_graph_optimiser.py by Fredrik Svedberg · 1 year, 11 months ago
  69. 25f48dd MLBEDSW-6314 Static optimisation for quantise OP by Ayaan Masood · 1 year, 11 months ago
  70. 4965fae MLBEDSW-6313 Static optimisation for Shape OP by Ayaan Masood · 1 year, 11 months ago
  71. a2ec5aa MLBEDSW-5384 FC layers run on NPU if underlying shape is 2D by Ayaan Masood · 2 years, 1 month ago
  72. f9267da MLBEDSW-6407: Vela fails with TypeError in npu_performance by Tim Hall · 2 years, 1 month ago
  73. d857507 Update version of Black to 22.3.0 by Jonas Ohlsson · 2 years, 2 months ago
  74. 3c5cfe9 MLBEDSW-6298: MLCE: Unable to find a valid block config by Tim Hall · 2 years, 2 months ago
  75. 43d2758 MLBEDSW-5332: Bug fix optimise_strided_conv by Louis Verhaard · 2 years, 2 months ago
  76. 1ddb2ed Vela: Fix diff in mean op by Diqing Zhong · 2 years, 2 months ago
  77. 0ab2edc MLBEDSW-6247: MLCE: Issue when running a model with Padding by Tim Hall · 2 years, 3 months ago
  78. e546def MLBEDSW-3623: Diff on semantic_segmentation by Rickard Bolin · 2 years, 4 months ago
  79. 51a8dce MLBEDSW-5844: Inconsistent calculation of read shapes by Tim Hall · 2 years, 5 months ago
  80. 7d7cb67 MLBEDSW-5554: Place MEAN op exceeding max height with axis==1 on CPU by Rickard Bolin · 2 years, 5 months ago
  81. cc8569f MLBEDSW-5209 Vela: output diff depthwise with non-zero zero points by Fredrik Svedberg · 2 years, 7 months ago
  82. 3e13434 MLBEDSW-5440 MLCE: Vela crashing due to mismatch in padding shape in OFM by James Ward · 2 years, 7 months ago
  83. 7c0607e MLBEDSW-5315 MLCE: Vela to handle skip Tensor by Fredrik Svedberg · 2 years, 7 months ago
  84. f436ada TOSA: Support for TABLE operator (int8) by Patrik Gustavsson · 2 years, 8 months ago
  85. 0957e3e MLBEDSW-5102 Update removal of memory only operators by Jonas Ohlsson · 2 years, 9 months ago
  86. 1a7527c MLBEDSW-5052 Fix RELU scaling by Fredrik Svedberg · 2 years, 8 months ago
  87. f1580f0 TOSA: Added RESHAPE, SLICE and CONCAT by Patrik Gustavsson · 2 years, 9 months ago
  88. df99510 TOSA: Added Depthwise support by Patrik Gustavsson · 2 years, 9 months ago
  89. cce872b MLBEDSW-5056 Output diff wav2letter (int16) by Fredrik Svedberg · 2 years, 9 months ago
  90. c74682c TOSA: Support for AVGPOOL, MAXPOOL and CONV2D by Patrik Gustavsson · 2 years, 9 months ago
  91. fbfd96e Handle sg input and output for Squeeze operator by Jonas Ohlsson · 2 years, 9 months ago 3.1.0.rc2
  92. 81942e9 MLBEDSW-4913 Fix inception_v1/v3 output diff by Jonas Ohlsson · 2 years, 9 months ago
  93. 8ae2929 MLBEDSW-4974: create_const_tensor invalid argument by Tim Hall · 2 years, 10 months ago
  94. 45e653d MLBEDSW-4853: Refactor supported operators by Jonas Ohlsson · 2 years, 10 months ago
  95. 7519d50 MLBEDSW-4892: Fix crash affecting biases without quantization. by James Peet · 2 years, 10 months ago
  96. 8f1f9aa MLBEDSW-4838 Added basic TOSA support. by Patrik Gustavsson · 2 years, 11 months ago