Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 4, 2022

• Forward fix land race (unexpected successes) (#86186)

  Show description for b95e0fc
  

  Chillee

  authored and

  pytorchmergebot

  committed
  b95e0fc

  Copy full SHA for b95e0fc

• Symbolic shapes mega merge PR (Oct 3) (#86160)

  Show description for 79dd621
  

  ezyang

  authored and

  pytorchmergebot

  committed
  79dd621

  Copy full SHA for 79dd621

• Symintified factory functions (#86067)

  Show description for de75274
  

  Chillee

  authored and

  pytorchmergebot

  committed
  de75274

  Copy full SHA for de75274

• Batch of symintifications to allow more models to pass in inference (#86104)

  Show description for 82d9592
  

  Chillee

  authored and

  pytorchmergebot

  committed
  82d9592

  Copy full SHA for 82d9592

• Stop modifying the global logger on `import functorch` (#86147)

  Show description for a4ff07f
  

  zou3519

  authored and

  pytorchmergebot

  committed
  a4ff07f

  Copy full SHA for a4ff07f

• Require bias to be contiguous for depthwise3x3_winograd backend (#85711)

  Show description for fe19007
  

  soulitzer

  authored and

  pytorchmergebot

  committed
  fe19007

  Copy full SHA for fe19007

• [maskedtensor] use masked_softmax for forward/backward instead of regular softmax (#85845)

  Show description for bc1d884
  

  george-qi

  authored and

  pytorchmergebot

  committed
  bc1d884

  Copy full SHA for bc1d884

Commits on Oct 3, 2022

• .github: Improve sanity check for generated files (#86143)

  Show description for 0db9419
  

  seemethere

  authored and

  pytorchmergebot

  committed
  0db9419

  Copy full SHA for 0db9419

• [GHF] Make EasyCLA unskippable (#86161)

  Show description for 5ca0f9e
  

  malfet

  authored and

  pytorchmergebot

  committed
  5ca0f9e

  Copy full SHA for 5ca0f9e

• Unconditionally register Python decomps to Meta key in Python Dispatcher (#85750)

  Show description for f3d7ab5
  

  ezyang

  authored and

  pytorchmergebot

  committed
  f3d7ab5

  Copy full SHA for f3d7ab5

• Revert "Disable XLA test (#86123)" (#86151)

  Show description for 06ddb1c
  

  huydhn

  authored and

  pytorchmergebot

  committed
  06ddb1c

  Copy full SHA for 06ddb1c

• Switch to checking EasyCLA on merge (#86127)

  Show description for cda815d
  

  zpao

  authored and

  pytorchmergebot

  committed
  cda815d

  Copy full SHA for cda815d

• ANTIALIAS updated to Resampling.LANCZOS in torch/utils/tensorboard/summary.py (#85679)

  Show description for dfde7cf
  

  GithubUserAccountAmazing

  authored and

  pytorchmergebot

  committed
  dfde7cf

  Copy full SHA for dfde7cf

• [easy] fix nested view call taking in more than one -1 (#86134)

  Show description for 2494c31
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  2494c31

  Copy full SHA for 2494c31

• MPS: Add multinomial op (#80760)

  Show description for 6a842e3
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  6a842e3

  Copy full SHA for 6a842e3

• Added _unsafe_view decomp (#86103)

  Show description for 37013bb
  

  Chillee

  authored and

  pytorchmergebot

  committed
  37013bb

  Copy full SHA for 37013bb

• [MPS] Cast dot inputs to int32 when needed (#86140)

  Show description for 40a8cc2
  

  abhudev

  authored and

  pytorchmergebot

  committed
  40a8cc2

  Copy full SHA for 40a8cc2

• Correctly error if you pass in tensors where size arguments expected (#86126)

  Show description for 954660a
  

  ezyang

  authored and

  pytorchmergebot

  committed
  954660a

  Copy full SHA for 954660a

• Symintified all functions, not including factory functions (#86078)

  Show description for 2aa9e07
  

  ezyang

  authored and

  pytorchmergebot

  committed
  2aa9e07

  Copy full SHA for 2aa9e07

• Decay integer-only (Optional)SymIntArrayRef to IntList in IValue (#86094)

  Show description for cb87983
  

  ezyang

  authored and

  pytorchmergebot

  committed
  cb87983

  Copy full SHA for cb87983

• Preserve/strip OptionalSymIntArrayRef when finding real schema (#86114)

  Show description for 146db41
  

  ezyang

  authored and

  pytorchmergebot

  committed
  146db41

  Copy full SHA for 146db41

• Fix compile error for vs2022 #79358 (#85958)

  Show description for 1da7492
  

  waruqi

  authored and

  pytorchmergebot

  committed
  1da7492

  Copy full SHA for 1da7492

• [ONNX] Remove registration in __init__ (#86130)

  Show description for 36634d7
  

  justinchuby

  authored and

  pytorchmergebot

  committed
  36634d7

  Copy full SHA for 36634d7

• Re-introduce the functorch docs build (#85838) (#86125)

  Show description for e01d616
  

  zou3519

  authored and

  pytorchmergebot

  committed
  e01d616

  Copy full SHA for e01d616

• [MPS] Improve memory usage and performance utilizing garbage collector and adaptive commit (#86119)

  Show description for 75c0e3a
  

  razarmehr

  authored and

  pytorchmergebot

  committed
  75c0e3a

  Copy full SHA for 75c0e3a

• [MPS] Handle compatible inputs to where (#85946)

  Show description for 8860e48
  

  abhudev

  authored and

  pytorchmergebot

  committed
  8860e48

  Copy full SHA for 8860e48

• [GHF] Add commit statuses to checkruns conclusions (#86129)

  Show description for 2f69223
  

  malfet

  authored and

  pytorchmergebot

  committed
  2f69223

  Copy full SHA for 2f69223

• Custom sdp implementations dense (#85984)

  Show description for cd64776
  

  drisspg

  authored and

  pytorchmergebot

  committed
  cd64776

  Copy full SHA for cd64776

• [skip-ci] Fixed bad link in build_ci_governance.rst (#85933)

  Show description for 8d9472d
  

  vfdev-5

  authored and

  pytorchmergebot

  committed
  8d9472d

  Copy full SHA for 8d9472d

• [docs] Add `torch.channels_last_3d (#85888)

  Show description for 85d520d
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  85d520d

  Copy full SHA for 85d520d

• [FSDP] Delay moving tensor to CPU until necessary for optim_state_dict() (#85761)

  Show description for 2067b76
  

  fegin

  authored and

  pytorchmergebot

  committed
  2067b76

  Copy full SHA for 2067b76

• Revert "Require bias to be contiguous for depthwise3x3_winograd backend (#85711)"

  Show description for e23cede
  pytorchmergebot

  committed
  e23cede

  Copy full SHA for e23cede

• Update dist.scatter() documentation (#86069)

  Show description for c670bad
  

  zerolinux5

  authored and

  pytorchmergebot

  committed
  c670bad

  Copy full SHA for c670bad

• implementation of qmul using xnnpack (#86040)

  Show description for 2403d0c
  

  Cuiqing Li

  authored and

  pytorchmergebot

  committed
  2403d0c

  Copy full SHA for 2403d0c

• parallelize at file granularity (#85770)

  Show description for 7941b04
  

  clee2000

  authored and

  pytorchmergebot

  committed
  7941b04

  Copy full SHA for 7941b04

Pagination

Previous

Next