Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 31, 2024

• Unskip test_complex_type_conversions (pytorch#118694)

  Show description for 56718ca
  

  kit1980

  authored and

  pytorchmergebot

  committed
  56718ca

  Copy full SHA for 56718ca

• Add --filter-rank to torchrun: allow logs filtering by rank (pytorch#118562)

  Show description for 73229b4
  

  xmfan

  authored and

  pytorchmergebot

  committed
  73229b4

  Copy full SHA for 73229b4

• Add Silu to Dtensor Pointwise ops (pytorch#118702)

  Show description for 995f696
  

  drisspg

  authored and

  pytorchmergebot

  committed
  995f696

  Copy full SHA for 995f696

• Fix admm over empty tensors and broadcastable input (pytorch#118619)

  Show description for 74f4947
  

  malfet

  authored and

  pytorchmergebot

  committed
  74f4947

  Copy full SHA for 74f4947

• [export] Enforce serialization BC/FC with updater script. (pytorch#118424)

  Show description for 2d37a04
  

  zhxchen17

  authored and

  pytorchmergebot

  committed
  2d37a04

  Copy full SHA for 2d37a04

• Preliminary DeviceMesh + native c10d functional integration (pytorch#118423)

  Show description for 697ca4f
  

  yifuwang

  authored and

  pytorchmergebot

  committed
  697ca4f

  Copy full SHA for 697ca4f

• [FSDP2] Added initial _lazy_init and FQNs for debugging (pytorch#117881)

  Show description for e3cde68
  

  awgu

  authored and

  pytorchmergebot

  committed
  e3cde68

  Copy full SHA for e3cde68

• [vision hash update] update the pinned vision hash (pytorch#118700)

  Show description for f7ae454
  

  pytorchupdatebot

  authored and

  pytorchmergebot

  committed
  f7ae454

  Copy full SHA for f7ae454

• [audio hash update] update the pinned audio hash (pytorch#118699)

  Show description for 6d7cfb5
  

  pytorchupdatebot

  authored and

  pytorchmergebot

  committed
  6d7cfb5

  Copy full SHA for 6d7cfb5

• [PT-Vulkan] aten::conv1d - support any stride, padding, dilation (pytorch#118660)

  Show description for 0a7e2ce
  

  jorgep31415

  authored and

  pytorchmergebot

  committed
  0a7e2ce

  Copy full SHA for 0a7e2ce

• [lint] remove merge_base_with from .lintrunner.toml (pytorch#118677)

  Show description for 68a75d4
  

  suo

  authored and

  pytorchmergebot

  committed
  68a75d4

  Copy full SHA for 68a75d4

• [FSDP2] Sharded parameter in FSDPParam (pytorch#117877)

  Show description for 07a7fec
  

  awgu

  authored and

  pytorchmergebot

  committed
  07a7fec

  Copy full SHA for 07a7fec

• Enable nested namespace check in clang-tidy (pytorch#118506)

  Show description for 4a01904
  

  cyyever

  authored and

  pytorchmergebot

  committed
  4a01904

  Copy full SHA for 4a01904

• [meta registration] fix _efficient_attention_forward for jagged inputs (pytorch#118657)

  Show description for 1b03423
  

  davidberard98

  authored and

  pytorchmergebot

  committed
  1b03423

  Copy full SHA for 1b03423

• Reland: [aotinductor] Replicate split_cat from torch IR to predispatch IR" (pytorch#118590)

  Show description for 6fa162e
  

  frank-wei

  authored and

  pytorchmergebot

  committed
  6fa162e

  Copy full SHA for 6fa162e

Commits on Jan 30, 2024

• Fix a bug with python lambda capture (pytorch#118676)

  Show description for 7761ceb
  

  oulgen

  authored and

  pytorchmergebot

  committed
  7761ceb

  Copy full SHA for 7761ceb

• add torch.float64 precision support to the transformer test suite in TP/SP (pytorch#116436)

  Show description for 616e9db
  

  tianyu-l

  authored and

  pytorchmergebot

  committed
  616e9db

  Copy full SHA for 616e9db

• Flix lint after pytorch#117814 (pytorch#118689)

  Show description for 1f376b3
  

  atalman

  authored and

  pytorchmergebot

  committed
  1f376b3

  Copy full SHA for 1f376b3

• Fix/Temporarily disable tests broken due to triton version mismatch (pytorch#118661)

  Show description for 1e78dc9
  

  oulgen

  authored and

  pytorchmergebot

  committed
  1e78dc9

  Copy full SHA for 1e78dc9

• register decomposition for rsub in torch._refs (pytorch#118288)

  Show description for 2f7839e
  

  isuruf

  authored and

  pytorchmergebot

  committed
  2f7839e

  Copy full SHA for 2f7839e

• Fix signatures of torch.{add, sub, mul} (pytorch#118398)

  Show description for 04ded13
  

  isuruf

  authored and

  pytorchmergebot

  committed
  04ded13

  Copy full SHA for 04ded13

• [FSDP2] Added initial FSDPParamGroup, FSDPParam, ParamModuleInfo (pytorch#117867)

  Show description for 6ea233a
  

  awgu

  authored and

  pytorchmergebot

  committed
  6ea233a

  Copy full SHA for 6ea233a

• [FSDP2] Added mesh arg, FSDPState, move to device (pytorch#117814)

  Show description for ae6233e
  

  awgu

  authored and

  pytorchmergebot

  committed
  ae6233e

  Copy full SHA for ae6233e

• [FSDP2][Reland] Introduced initial fully_shard frontend (pytorch#118525)

  Show description for 7aa4b35
  

  awgu

  authored and

  pytorchmergebot

  committed
  7aa4b35

  Copy full SHA for 7aa4b35

• Fix missing permission in create release workflow (pytorch#118681)

  Show description for 48f8761
  

  huydhn

  authored and

  pytorchmergebot

  committed
  48f8761

  Copy full SHA for 48f8761

• Dont fuse write into read if indexing differs (pytorch#118210)

  Show description for 1aa836f
  

  eellison

  authored and

  pytorchmergebot

  committed
  1aa836f

  Copy full SHA for 1aa836f

• [quant][bc-breaking] Turn on fold_quantize by default (pytorch#118605)

  Show description for 82a7460
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  82a7460

  Copy full SHA for 82a7460

• Remove voznesenskym from the list of autoreviewers (pytorch#118680)

  Show description for ba1be17
  

  izaitsevfb

  authored and

  pytorchmergebot

  committed
  ba1be17

  Copy full SHA for ba1be17

• Workaround for super() calls in test_torchinductor_dynamic_shapes (pytorch#118586)

  Show description for f2682e7
  

  clee2000

  authored and

  pytorchmergebot

  committed
  f2682e7

  Copy full SHA for f2682e7

• Enable possibly-undefined error code (pytorch#118533)

  Show description for 4f5785b
  

  clee2000

  authored and

  pytorchmergebot

  committed
  4f5785b

  Copy full SHA for 4f5785b

• [inductor] Use at::detail::empty_strided_* in cpp_wraper mode (pytorch#118490)

  Show description for e332653
  

  jansel

  authored and

  pytorchmergebot

  committed
  e332653

  Copy full SHA for e332653

• [BE]: Apply RUF025 dict.fromkeys preview rule (pytorch#118637)

  Show description for 1562dae
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  1562dae

  Copy full SHA for 1562dae

• Add separate logging target for cudagraphs (pytorch#118329)

  Show description for e33e88e
  

  eellison

  authored and

  pytorchmergebot

  committed
  e33e88e

  Copy full SHA for e33e88e

• [c10d] Log the last enqueued and completed collective (pytorch#118582)

  Show description for e180218
  

  shuqiangzhang

  authored and

  pytorchmergebot

  committed
  e180218

  Copy full SHA for e180218

• [PT-Vulkan] aten::unsqueeze - nit optimization (pytorch#118575)

  Show description for 9247641
  

  jorgep31415

  authored and

  pytorchmergebot

  committed
  9247641

  Copy full SHA for 9247641

Pagination

Previous

Next