Commits

Branch selector

 111c86b

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 10, 2023

• Revert "[CI] Move M1 testing to periodic (pytorch#94608)"

  Show description for 111c86b
  pytorchmergebot

  committed
  111c86b

  Copy full SHA for 111c86b

• [MPS] Fix the crash in huberloss with Float16 (pytorch#94567)

  Show description for 7c4acda
  

  razarmehr

  authored and

  pytorchmergebot

  committed
  7c4acda

  Copy full SHA for 7c4acda

• Continue support sharding pipes in tud.datapipes.iter.grouping as deprecated (pytorch#94527)

  Show description for d8f4026
  

  wenleix

  authored and

  pytorchmergebot

  committed
  d8f4026

  Copy full SHA for d8f4026

• [CI] Move M1 testing to periodic (pytorch#94608)

  Show description for 5c16788
  

  malfet

  authored and

  pytorchmergebot

  committed
  5c16788

  Copy full SHA for 5c16788

• Run test_torchinductor*.py with implicit_fallbacks=False (pytorch#94039)

  Show description for e116ca9
  

  Fabio Rocha

  authored and

  pytorchmergebot

  committed
  e116ca9

  Copy full SHA for e116ca9

• Pass input tensor __dict__ along to placeholder nodes (pytorch#94080)

  Show description for e44586a
  

  voznesenskym

  authored and

  pytorchmergebot

  committed
  e44586a

  Copy full SHA for e44586a

• [BE] Modernize PyTorch even more for 3.8 with pyupgrade (pytorch#94520)

  Show description for 9171f7d
  

  Skylion007

  authored and

  pytorchmergebot

  committed
  9171f7d

  Copy full SHA for 9171f7d

• [SDPA] update type hint for scaled_dot_product_attention and documentation (pytorch#94008)

  Show description for 70026aa
  

  drisspg

  authored and

  pytorchmergebot

  committed
  70026aa

  Copy full SHA for 70026aa

• Fix div by fp64 scalar issue on xla device (pytorch#94459)

  Show description for 9bef1eb
  

  ymwangg

  authored and

  pytorchmergebot

  committed
  9bef1eb

  Copy full SHA for 9bef1eb

• Cleaning up some logic in tools/shared/cwrap_common.py (pytorch#94475)

  Show description for 67513ae
  

  joesho112358

  authored and

  pytorchmergebot

  committed
  67513ae

  Copy full SHA for 67513ae

• add compile reason in InstructionTranslator RETURN_VALUE (pytorch#94176) (pytorch#94367)

  Show description for 51cec7b
  

  zjjott

  authored and

  pytorchmergebot

  committed
  51cec7b

  Copy full SHA for 51cec7b

• [MPS] Fix cumsum for integral data types (pytorch#94530)

  Show description for 92d8c4b
  

  DenisVieriu97

  authored and

  pytorchmergebot

  committed
  92d8c4b

  Copy full SHA for 92d8c4b

• [fx] Fix matching args (pytorch#94375)

  Show description for d990dda
  

  angelayi

  authored and

  pytorchmergebot

  committed
  d990dda

  Copy full SHA for d990dda

• fix: forbid multi-index for index_select over scalar (pytorch#94347)

  Show description for db6cfff
  

  ganler

  authored and

  pytorchmergebot

  committed
  db6cfff

  Copy full SHA for db6cfff

• feature: adding the ability to restore shapes after loading a traced model (pytorch#90744)

  Show description for 0d0ebcd
  

  mnuyens

  authored and

  pytorchmergebot

  committed
  0d0ebcd

  Copy full SHA for 0d0ebcd

• Fix bug in unsqueeze_nested stride calculation (pytorch#88688)

  Show description for c7c7238
  

  mikaylagawarecki

  authored and

  pytorchmergebot

  committed
  c7c7238

  Copy full SHA for c7c7238

• [ONNX] Skip import test for experimental files (pytorch#94552)

  Show description for 889a464
  

  BowenBao

  authored and

  pytorchmergebot

  committed
  889a464

  Copy full SHA for 889a464

• port sparse_mm.reduce to pytorch and optimize it on CPU (pytorch#83727)

  Show description for c620ece
  

  mingfeima

  authored and

  pytorchmergebot

  committed
  c620ece

  Copy full SHA for c620ece

• Add config option to reduce warnings in inductor (pytorch#94413)

  Show description for 24ae50b
  

  jansel

  authored and

  pytorchmergebot

  committed
  24ae50b

  Copy full SHA for 24ae50b

• [MPS] Fix min/max_reduction_with_dim ops (pytorch#94386)

  Show description for 1d39806
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  1d39806

  Copy full SHA for 1d39806

• [MPS] Add im2col and col2im to Fallback (pytorch#94491)

  Show description for 0fe1158
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  0fe1158

  Copy full SHA for 0fe1158

• WelfordOps: Remove combine_t and use acc_scalar_t instead (pytorch#94522)

  Show description for a21bddc
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  a21bddc

  Copy full SHA for a21bddc

• [decomp] Use var_mean in native_batch_norm decomposition (pytorch#94140)

  Show description for e22e323
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  e22e323

  Copy full SHA for e22e323

• Fix embedding_dense_backward to not cast indiices to floats (pytorch#94572)

  Show description for e844120
  

  Chillee

  authored and

  pytorchmergebot

  committed
  e844120

  Copy full SHA for e844120

• Don't throw tf32 warning if no nodes in graph are matmuls + fp32 + cuda (pytorch#94561)

  Show description for 1770ccf
  

  Chillee

  authored and

  pytorchmergebot

  committed
  1770ccf

  Copy full SHA for 1770ccf

• Revert "update aten op overload to not use from to avoid compile errors (pytorch#89797)"

  Show description for f152a79
  pytorchmergebot

  committed
  f152a79

  Copy full SHA for f152a79

• teach inductor to handle floor (pytorch#94341)

  Show description for a5daea6
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  a5daea6

  Copy full SHA for a5daea6

• inductor: don't do transpose vectoriztion if input ld depends on most inner var (pytorch#94493)

  Show description for 02b8a7f
  

  XiaobingSuper

  authored and

  pytorchmergebot

  committed
  02b8a7f

  Copy full SHA for 02b8a7f

• Renamed passes to options in torch.compile (pytorch#94500)

  Show description for 3a12b16
  

  Chillee

  authored and

  pytorchmergebot

  committed
  3a12b16

  Copy full SHA for 3a12b16

• [MPS] Fix the Channels last bug with GradientWithInput. (pytorch#94384)

  Show description for 59e8756
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  59e8756

  Copy full SHA for 59e8756

• [MPS] Casting int64 to int32 for reduction ops and raise warning. (pytorch#94484)

  Show description for 8dbe63c
  

  kulinseth

  authored and

  pytorchmergebot

  committed
  8dbe63c

  Copy full SHA for 8dbe63c

• don't call floor for symint unless necessary (pytorch#94365)

  Show description for 715f373
  

  Natalia Gimelshein

  authored and

  pytorchmergebot

  committed
  715f373

  Copy full SHA for 715f373

• Enable Python-3.11 binary builds across the board (pytorch#94430)

  Show description for 89df0e4
  

  malfet

  authored and

  pytorchmergebot

  committed
  89df0e4

  Copy full SHA for 89df0e4

• [MPS] Fix batchnorm forward and backward pass (pytorch#94351)

  Show description for a1f15fb
  

  DenisVieriu97

  authored and

  pytorchmergebot

  committed
  a1f15fb

  Copy full SHA for a1f15fb

• [MPS] Fix addmm calculation (pytorch#94534)

  Show description for 2ad2900
  

  DenisVieriu97

  authored and

  pytorchmergebot

  committed
  2ad2900

  Copy full SHA for 2ad2900

Pagination

Previous

Next