Google Git
Sign in
android / platform / external / pytorch / HEAD
  1. ab2ddb1 Empty merge ab/12770256 into aosp-main-future by Xin Li · 12 months ago main
  2. 0f68a92 Add janitors to the OWNERS file am: 9b654ccb26 by Sadaf Ebrahimi · 12 months ago
  3. 9b654cc Add janitors to the OWNERS file by Sadaf Ebrahimi · 12 months ago
  4. 9ea4d19 Upgrade pytorch to v2.5.1 am: f6beb24584 am: db6a337ffc by Qiao Li · 1 year, 1 month ago aml_ase_351412000 aml_doc_351311000 aml_doc_351412000 aml_mpr_351311140 aml_mpr_351416060 aml_per_351410020 aml_rkp_351310000 aml_sdk_351415000 aml_sta_351311000
  5. db6a337 Upgrade pytorch to v2.5.1 am: f6beb24584 by Qiao Li · 1 year, 1 month ago
  6. f6beb24 Upgrade pytorch to v2.5.1 by Qiao Li · 1 year, 1 month ago
  7. fc756f5 Python soong target not support "apex_available" am: b5ea807cbd am: 5f9952124c by Qiao Li · 1 year, 2 months ago aml_cbr_351310040 aml_ips_351310000 aml_net_351310100 aml_uwb_351310040 aml_wif_351310040
  8. 765c786 Merge remote-tracking branch 'origin/upstream' am: 6fa82c6c96 am: 3f809a3fcf by Inna Palant · 1 year, 2 months ago
  9. 5f99521 Python soong target not support "apex_available" am: b5ea807cbd by Qiao Li · 1 year, 2 months ago
  10. 3f809a3 Merge remote-tracking branch 'origin/upstream' am: 6fa82c6c96 by Inna Palant · 1 year, 2 months ago
  11. b5ea807 Python soong target not support "apex_available" by Qiao Li · 1 year, 2 months ago
  12. 6fa82c6 Merge remote-tracking branch 'origin/upstream' by Inna Palant · 1 year, 2 months ago
  13. a8d6afb Disabling amp context when invoking compiler (#138659) by pytorchbot · 1 year, 2 months ago upstream-release/2.5 upstream/v2.5.1 upstream/v2.5.1-rc1
  14. f31b8bb [MPS] Fix sliced cast (#138535) by pytorchbot · 1 year, 2 months ago
  15. 848e7ac [SDPA-CUDNN] Make CuDNN Attention Opt in (#138587) by pytorchbot · 1 year, 2 months ago
  16. 885c823 Update doc copyrights to 2024 (#138650) by pytorchbot · 1 year, 2 months ago
  17. 8c3ed97 Update cpuinfo submodule (#138600) by Nikita Shulga · 1 year, 2 months ago
  18. e3a1028 Include platform/external/pytorch by Emily Fu · 1 year, 2 months ago upstream
  19. 70cf2bb Add link to torch.compile the missing manual in troubleshooting (#137369) by pytorchbot · 1 year, 2 months ago
  20. cde6b38 Don't try to load cufile (#138539) by pytorchbot · 1 year, 2 months ago
  21. 4076a73 [Cherry-Pick] Use cuda 12.4 pytorch_extra_install_requirements as default (#138526) by Andrey Talman · 1 year, 2 months ago
  22. a97c151 update getting started xpu (#138090) by pytorchbot · 1 year, 2 months ago
  23. 9a0dfa6 Initial empty repository by Inna Palant (xWF) · 1 year, 2 months ago
  24. 32f585d [Release only] use triton 3.1.x from pypi (#137895) by Andrey Talman · 1 year, 2 months ago upstream/v2.5.0 upstream/v2.5.0-rc10
  25. 417a076 [split build] move periodic split builds into own concurrency group (#135510) (#137265) by Andrey Talman · 1 year, 2 months ago upstream/v2.5.0-rc9
  26. 119e734 [RELEASE-ONLY CHANGES] Fix dependency on filesystem on Linux (#137242) by Sergii Dymchenko · 1 year, 2 months ago upstream/v2.5.0-rc8
  27. 783a6a4 [MPS] Add regression test for `fft.fftfreq` (#137215) by pytorchbot · 1 year, 2 months ago
  28. 5375201 [MPS] Add missing dispatch to rshift.Tensor (#137212) by pytorchbot · 1 year, 2 months ago
  29. 1de132e [MPS] Fix 5D+ reductions over negative dimentions (#137211) by pytorchbot · 1 year, 2 months ago
  30. 0b1b609 [NCCL] Don't override `waitUntilInitialized`'s setting of `comm->initialized_` (#137210) by pytorchbot · 1 year, 2 months ago
  31. 0b45af9 Fix addmm silent correctness on aarch64 (#137208) by pytorchbot · 1 year, 2 months ago
  32. 1a0b166 [ONNX] Add assertion nodes to ignoring list (#137214) by pytorchbot · 1 year, 2 months ago
  33. 3a541ef Clarify that `libtorch` API is C++17 compatible (#137206) by pytorchbot · 1 year, 2 months ago
  34. f8c4c25 [Release only] Set WITH_PUSH when WITH_PUSH_ROCM is set (#137177) by Andrey Talman · 1 year, 2 months ago
  35. 8af31b2 [RELEASE-ONLY Change] Push ROCm images on RC (#137148) by Sergii Dymchenko · 1 year, 2 months ago upstream/v2.5.0-rc7
  36. 8a71edc [RELEASE-ONLY CHANGES] Disable slow workflows (#136805) by Sergii Dymchenko · 1 year, 2 months ago
  37. 058d3de [dynamo] Do not treat user defined nn module attributes static for dynamic shape infra (#137025) by Animesh Jain · 1 year, 2 months ago
  38. 17d2589 [FlexAttention] Fix output layout (#135882) (#136905) by Driss Guessous · 1 year, 2 months ago
  39. 70298e9 [Cherry-pick][DSD] Fix distributed state dict full_state_dict option hang during set_state_dict (#135725) and Fix loading uneven full tensor into sharded state dict (#136365) (#136903) by Iris Z · 1 year, 2 months ago
  40. 69ed7c7 [SymmetricMemory] improve multicast initialization/fallback logic (#136894) by Yifu Wang · 1 year, 2 months ago upstream/v2.5.0-rc6
  41. d80f521 fix requirements.txt installation failure issue on Windows (#136893) by Jing Xu · 1 year, 2 months ago
  42. 57717c8 Fix lint (#137052) by pytorchbot · 1 year, 2 months ago
  43. 550ed97 [cuDNN][SDPA] cherrypick Support `attn_bias` in cuDNN (#130482) (#136885) by eqy · 1 year, 2 months ago
  44. 051df20 [ROCm] Update to AOTriton 0.7b (Cherry-picked) (#135869) by Xinya Zhang · 1 year, 2 months ago
  45. bc421d4 [RELEASE ONLY CHANGES] Revert XNNPACK Update (#136522) by Gregory Comer · 1 year, 3 months ago
  46. aa574ab SDPA regression fix to work around high-precision by default (#136536) by Joel Schlosser · 1 year, 3 months ago
  47. 24bd87d [Docs] fix inconsistent docs in conv1d, conv2d, and conv3d (#136813) by pytorchbot · 1 year, 3 months ago upstream/v2.5.0-rc5
  48. 6101aaf [Update] Update note for Getting Started with PyTorch on Intel GPUs (#136731) by pytorchbot · 1 year, 3 months ago
  49. 396413f Fix ROCm skip decorator for test_ddp_tp and multiprocess UTs (#136161) (#136801) by Sergii Dymchenko · 1 year, 3 months ago
  50. c25781c Update current maintainers (#136769) by pytorchbot · 1 year, 3 months ago
  51. ecd3306 Constraint setuptools to 72.1.0 or older in requirements.txt (#136729) by pytorchbot · 1 year, 3 months ago
  52. 1715708 Revert "Trace fwd graph under no_grad mode #134872" (#136734) by Yanbo Liang · 1 year, 3 months ago
  53. 2e2c00f Make test_skip_data_serialization regex more flexible (#136710) by pytorchbot · 1 year, 3 months ago
  54. cbe476a Disable iOS workflow (#136706) by pytorchbot · 1 year, 3 months ago
  55. 4b030d4 [RELEASE-ONLY CHANGES] Don't push to https://ghcr.io/ (#136703) by Sergii Dymchenko · 1 year, 3 months ago
  56. 9b80dde Fix hardcoded ROCm paths in `Caffe2Targets.cmake` (#136700) by pytorchbot · 1 year, 3 months ago
  57. 6e86793 [ROCm] upgrade ROCm CI builds to py3.10 (#134108) (#136696) by Sergii Dymchenko · 1 year, 3 months ago
  58. 7c550fe [ROCm][CI] upgrade CI to ROCm 6.2 (#132555) (#136467) by Jeff Daily · 1 year, 3 months ago upstream/v2.5.0-rc4
  59. 7a00785 [ROCm] Cherry-pick unit test fixes to release/2.5 (#136557) by pytorchbot · 1 year, 3 months ago
  60. 4e6a99e fix stride compare failed when size value equal to one in ForeachUtils.h (#136426) by Shan19900305 · 1 year, 3 months ago
  61. dd73223 [ROCm] [BUGFIX] Re-enable rocm-specific tuning parameters v2 (#133852) (#136139) by Jack Taylor · 1 year, 3 months ago
  62. c5e5254 Fix xpu memory stats error (#135818) (#136420) by Yu, Guangye · 1 year, 3 months ago
  63. ffed7b7 Fix dynamo benchmark skip logic for cpu device (#135193) (#135793) by xiangdong · 1 year, 3 months ago
  64. becdf8a [Inductor] Increase multiplier to 3 for Inductor AMP FP16 benchmark correctness check (#135932) (#136262) by jiayisunx · 1 year, 3 months ago
  65. fb276d2 [ONNX] Fix numpy method to return the correct type (#136162) (#136203) by Justin Chu · 1 year, 3 months ago
  66. b7de793 Update torch-xpu-ops pin (ATen XPU implementation) (#135833) by Feng Yuan · 1 year, 3 months ago
  67. 1954439 Update document for autocast on CPU (#136082) by Cao E · 1 year, 3 months ago
  68. 3920988 [Cherry-pick] [ONNX] Update fake mode usage in onnx docs (#135512) and Drop final None values as inputs for nodes in exporter graph (#135520) (#136005) by Shubham Bhokare · 1 year, 3 months ago
  69. 1db2a65 [DCP] Fixes the stateless optimizer issue of distributed state_dict (… (#136000) by Aaron Gokaslan · 1 year, 3 months ago
  70. 4b5bf41 [inductor] [cpp] fix the input contiguous check in max-autotune (#135561) by Chunyuan WU · 1 year, 3 months ago
  71. a889c85 [Inductor] simplify indexing_exprs in LoopBody._init_with_copy (#135574) (#135935) by jiayisunx · 1 year, 3 months ago
  72. 813e064 [ONNX] Fix symbolic values and numpy implementation (#135786) (#135868) by Justin Chu · 1 year, 3 months ago
  73. 9887030 Revert "[fx] Bypass custom __setattr__ in Node.__init__ (#135079)" (#… (#135625) by Sergii Dymchenko · 1 year, 3 months ago
  74. 9e315fe Revert "[Release only] Temporary disable triton xpu build" (#136276) by Andrey Talman · 1 year, 3 months ago
  75. 6b14e6c [Release only] Temporary disable triton xpu build (#136206) by Andrey Talman · 1 year, 3 months ago upstream/v2.5.0-rc3
  76. 828d686 [ONNX] Fix scaled_dot_product_attention with float scale (#135710) by Ti-Tai Wang · 1 year, 3 months ago upstream/v2.5.0-rc2
  77. 612fc7c [ONNX] Improves documentation of ONNX exporter (#135526) by Justin Chu · 1 year, 3 months ago
  78. cea5620 [Cherry-Pick] Bump triton pin and release version, revert temporary changes to build from pin (#135613) by Andrey Talman · 1 year, 3 months ago
  79. ba27502 Use [email protected] for create_release.yml (#135534) by pytorchbot · 1 year, 3 months ago
  80. 4a3dabd [RELEASE-ONLY CHANGES] Temp changes to build triton from pin for first RC (#135517) by Andrey Talman · 1 year, 3 months ago upstream/v2.5.0-rc1
  81. e130696 [RELEASE-ONLY CHANGES] Branch Cut for Release 2.5 (#135506) by Andrey Talman · 1 year, 3 months ago
  82. b7eb725 docs: `torch.nn.utils.rnn.pack_padded_sequence`: docs improve (#135417) by Alexander Kurakin · 1 year, 3 months ago
  83. c1ae78b [inductor] calibration inductor windows uts (18/N) (#135449) by Xu Han · 1 year, 3 months ago
  84. defb515 [NJT]Add permute ops support (#135336) by yuqingj · 1 year, 3 months ago
  85. 31c4e0d [inductor] Cleanup analysis done at lowering time (#135412) by Jason Ansel · 1 year, 3 months ago
  86. 53290ca [inductor] Refactor BaseSchedulerNode.__init__ (#135400) by Jason Ansel · 1 year, 3 months ago
  87. 16f5155 [inductor] Fast path for extract_read_writes without tracing (#135377) by Jason Ansel · 1 year, 3 months ago
  88. 37144be [inductor] Remove ReadWrites.op_counts (#135306) by Jason Ansel · 1 year, 3 months ago
  89. 3bdc54e [inductor] Refactor LoopBody.memory_usage (#135286) by Jason Ansel · 1 year, 3 months ago
  90. 2196f32 [22/N] Fix clang-tidy warnings in jit (#135319) by cyy · 1 year, 3 months ago
  91. cfc227a [reland][dtensor] move DTensor to public namespace (#134203) by Wanchao Liang · 1 year, 3 months ago
  92. 20cab91a [dynamo] Remove skip from jit freeze tests (#135281) by Animesh Jain · 1 year, 3 months ago
  93. a6fae2e Use BRGEMM for Half flash attention forward kernel (#131879) by CaoE · 1 year, 3 months ago
  94. 042f2f7 [ONNX] Re-raise the exception if the dynamic shapes cannot be refined (#135418) by Justin Chu · 1 year, 3 months ago
  95. fd494dd Change wrapped_linear_prepack and wrapped_quantized_linear_prepacked to private by adding _ as prefix (#135401) by Huamin Li · 1 year, 3 months ago
  96. 8334cb2 remove commented out breakpoints (#135363) by Bob Ren · 1 year, 3 months ago
  97. e72ed47 [Dynamo] Fix Huggingface PretrainedConfig get non const attr (#135413) by Yanbo Liang · 1 year, 3 months ago
  98. 3bebc09 [FlexAttention] Align the matmul tensorcore usage (#135168) by drisspg · 1 year, 3 months ago
  99. a2db22e [inductor] Catch BrokenProcessPool and print a more helpful message. (#135120) by Sam Larsen · 1 year, 3 months ago
  100. eac5e12 [inductor] Move LoopBody to its own file (#135257) by Jason Ansel · 1 year, 3 months ago