Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 14, 2022

• Remove cpu path from lobpcg's basis helper (pytorch#88984)

  Show description for b0c86ca
  

  IvanYashchuk

  authored and

  pytorchmergebot

  committed
  b0c86ca

• don't use prims.unsqueeze in group_norm (pytorch#88927)

  Show description for 06f1b52
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  06f1b52

• Run test_torchinductor_opinfo CPU tests if triton not installed (pytorch#88934)

  Show description for c8f3d1c
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  c8f3d1c

• reland "Do not use unsafe restriding for subclasses (pytorch#87610)" (pytorch#88343)

  Show description for ec4eada
  

  bdhirsh

  authored and

  pytorchmergebot

  committed
  ec4eada

• TorchDynamo: skip convolution fusion when convolution's padding is string (pytorch#88794)

  Show description for 9943d46
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  9943d46

• Fake Tensor For (ConvFusion) Propagation (pytorch#88414)

  Show description for 15ef066
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  15ef066

• Revert "Run test_torchinductor_opinfo CPU tests if triton not installed (pytorch#88934)"

  Show description for 5e6cefd
  pytorchmergebot

  committed
  5e6cefd

• Run test_torchinductor_opinfo CPU tests if triton not installed (pytorch#88934)

  Show description for 8371bb8
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  8371bb8

• TorchDynamo: Add convolution binary+unary fusion for cpu in inference mode (pytorch#88412)

  Show description for 072920c
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  072920c

• [xla hash update] update the pinned xla hash (pytorch#88982)

  Show description for cb4842c
  pytorchmergebot

  committed
  cb4842c

• Fix typos in messages under aten (pytorch#88964)

  Show description for 0329684
  

  kiszk

  authored and

  pytorchmergebot

  committed
  0329684

• TorchDynamo: Add convolution binary(inplace) fusion for cpu in inference mode (pytorch#88403)

  Show description for 4ad7b17
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  4ad7b17

• fix typo: AT_MKLDNN_EBABLED => AT_MKLDNN_ENABLED (pytorch#88952)

  Show description for 06486cd
  

  jinyouzhi

  authored and

  pytorchmergebot

  committed
  06486cd

• Revert "Symintify decomps for split and upsample_bilinear; Fix decomp for _softmax_backward_data and native_dropout_backward (pytorch#88761)"

  Show description for eea506a
  pytorchmergebot

  committed
  eea506a

Commits on Nov 13, 2022

• Fix: [ATen] Add some missing moves (pytorch#88514)

  Show description for 48dc24d
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  48dc24d

• Symintify decomps for split and upsample_bilinear; Fix decomp for _softmax_backward_data and native_dropout_backward (pytorch#88761)

  Show description for 9eabcc3
  

  SherlockNoMad

  authored and

  pytorchmergebot

  committed
  9eabcc3

• [primTorch] Add ref for complex (pytorch#88562)

  Show description for 76af714
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  76af714

• Skip dynamo benchmark tests under TSAN (pytorch#88895)

  Show description for 8f7e519
  

  jansel

  authored and

  pytorchmergebot

  committed
  8f7e519

• meta function for max_pool2d_with_indices_backward (pytorch#88743)

  Show description for 52be0c4
  

  anjali411

  authored and

  pytorchmergebot

  committed
  52be0c4

• Revert "[reland][dynamo] Better support for nn.Module (pytorch#88959)"

  Show description for 98bcb4a
  pytorchmergebot

  committed
  98bcb4a

• [reland][dynamo] fixes dict changed during runtime error (pytorch#88877)

  Show description for 897d029
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  897d029

• [Dynamo][FSDP] Migrate to ModuleWrapPolicy (pytorch#88453)

  Show description for 4284862
  

  awgu

  authored and

  pytorchmergebot

  committed
  4284862

• Move xnnpack taget to fb code base (pytorch#88909)

  Show description for bca75fd
  

  cccclai

  authored and

  pytorchmergebot

  committed
  bca75fd

• [dynamo] Skip frame when graph break in a loop (pytorch#88857)

  Show description for 2b12bfc
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  2b12bfc

• [reland][dynamo] Better support for nn.Module (pytorch#88959)

  Show description for e950afc
  

  anijain2305

  authored and

  pytorchmergebot

  committed
  e950afc

• [dynamo] Port all pytorch/dynamo and test/dynamo pieces over from symbolic-shapes branch (pytorch#88768)

  Show description for 06ce133
  

  voznesenskym

  authored and

  pytorchmergebot

  committed
  06ce133

• [FSDP] Fix FSDP.clip_grad_norm_() for NO_SHARD (pytorch#88955)

  Show description for 4f2639e
  

  awgu

  authored and

  pytorchmergebot

  committed
  4f2639e

• Fix XLA symbolic shapes binding (pytorch#88928)

  Show description for 46796fe
  

  ezyang

  authored and

  pytorchmergebot

  committed
  46796fe

• Vectorized CPU code implementing left shift operator. (pytorch#88607)

  Show description for 2aca97c
  

  alexsamardzic

  authored and

  pytorchmergebot

  committed
  2aca97c

Commits on Nov 12, 2022

• [16/N] Add _allgather_base custom op with CPU/CUDA implementation (pytorch#88889)

  Show description for df1df9d
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  df1df9d

• torchdynamo support self.modules() for nn_module (pytorch#88695)

  Show description for 3765621
  

  ydwu4

  authored and

  pytorchmergebot

  committed
  3765621

• Turn internal assert when saved tensor is detached inplace into torch check (pytorch#88860)

  Show description for 27dc03e
  

  soulitzer

  authored and

  pytorchmergebot

  committed
  27dc03e

• [primTorch] Improve narrow and narrow_copy: refs, tests, docs (pytorch#87045)

  Show description for 4270bb3
  

  nkaretnikov

  authored and

  pytorchmergebot

  committed
  4270bb3

• [15/N] Add allreduce_coalesced custom op with CPU/CUDA implementations (pytorch#88846)

  Show description for 6e5f736
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  6e5f736

• Revert "[dynamo][api] Better support of torch.nn.Module (pytorch#88629)"

  Show description for ae2c668
  pytorchmergebot

  committed
  ae2c668

Pagination

PreviousNext