Commits

Breadcrumbs

History for

onv1

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 29, 2020

• Backend Cleanroom port (#13)

  Show description for f4603e3
  MarisaKirisame

  committed
  f4603e3

  Copy full SHA for f4603e3

Commits on Sep 1, 2020

• [quant][graphmode][fx][refactor] Move patterns to separate files (#43891)

  Show description for d15b9d9
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  d15b9d9

  Copy full SHA for d15b9d9

• [FX] Better error when unpacking Proxy (#43740)

  Show description for 8d53df3
  

  James Reed

  authored and

  facebook-github-bot

  committed
  8d53df3

  Copy full SHA for 8d53df3

• Delete raise_from from torch._six (#43981)

  Show description for e49dd9f
  

  malfet

  authored and

  facebook-github-bot

  committed
  e49dd9f

  Copy full SHA for e49dd9f

• Enable TF32 support for cuDNN (#40737)

  Show description for 5e97f25
  

  zasdfgbnm

  authored and

  facebook-github-bot

  committed
  5e97f25

  Copy full SHA for 5e97f25

• Check SparseAdam params are dense on init (#41966) (#43668)

  Show description for 24eea36
  

  Randall Hunt

  authored and

  facebook-github-bot

  committed
  24eea36

  Copy full SHA for 24eea36

• [FX] Pickle serialization of GraphModule via forward source (#43674)

  Show description for a1a2366
  

  James Reed

  authored and

  facebook-github-bot

  committed
  a1a2366

  Copy full SHA for a1a2366

• [FX] Support tensor-valued constants (#43666)

  Show description for 73f7d63
  

  James Reed

  authored and

  facebook-github-bot

  committed
  73f7d63

  Copy full SHA for 73f7d63

• Reset DataLoader workers instead of creating new ones (#35795)

  Show description for 5472426
  

  Emilio Castillo

  authored and

  facebook-github-bot

  committed
  5472426

  Copy full SHA for 5472426

• Revert D23216393: Support work.result() to get result tensors for allreduce for Gloo, NCCL backends

  Show description for 7035cd0
  

  malfet

  authored and

  facebook-github-bot

  committed
  7035cd0

  Copy full SHA for 7035cd0

• Add typing annotations for torch.nn.quantized.dynamic.modules.rnn (#43186)

  Show description for 63a0bb0
  

  guilhermeleobas

  authored and

  facebook-github-bot

  committed
  63a0bb0

  Copy full SHA for 63a0bb0

• Error printing extension support for multiline errors (#43807)

  Show description for e3cb582
  

  Lilyjjo

  authored and

  facebook-github-bot

  committed
  e3cb582

  Copy full SHA for e3cb582

• Move Autograd to an alias dispatch key (#43070)

  Show description for 2242320
  

  Ailing Zhang

  authored and

  facebook-github-bot

  committed
  2242320

  Copy full SHA for 2242320

• MaxPool1d without indices optimization (#43745)

  Show description for 13a48ac
  

  heitorschueroff

  authored and

  facebook-github-bot

  committed
  13a48ac

  Copy full SHA for 13a48ac

• Support work.result() to get result tensors for allreduce for Gloo, NCCL backends (#43386)

  Show description for 0b2694c
  

  mrzzd

  authored and

  facebook-github-bot

  committed
  0b2694c

  Copy full SHA for 0b2694c

• Revert "Added support for Huber Loss (#37599)" (#43351)

  Show description for fab012a
  

  vincentqb

  authored and

  facebook-github-bot

  committed
  fab012a

  Copy full SHA for fab012a

• Fix NaN propagation in TE fuser's min/max implementation (#43609)

  Show description for c14a361
  

  bertmaher

  authored and

  facebook-github-bot

  committed
  c14a361

  Copy full SHA for c14a361

• [ONNX] Update slice symbolic function (#42935)

  Show description for 820c4b0
  

  KsenijaS

  authored and

  facebook-github-bot

  committed
  820c4b0

  Copy full SHA for 820c4b0

• Preserve python backtrace in autograd engine errors. (#43684)

  Show description for f1624b8
  

  pritamdamania

  authored and

  facebook-github-bot

  committed
  f1624b8

  Copy full SHA for f1624b8

• [reland][quant][graphmode][fx] Add support for weight prepack folding (#43728) (#43902)

  Show description for 825c109
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  825c109

  Copy full SHA for 825c109

• Update torch.range warning message regarding the removal version number (#43569)

  Show description for 6da26cf
  

  xuhdev

  authored and

  facebook-github-bot

  committed
  6da26cf

  Copy full SHA for 6da26cf

• Revert D23335106: [quant][graphmode][fix] Fix insert quant dequant for observers without qparams

  Show description for f229d2c
  

  Ailing Zhang

  authored and

  facebook-github-bot

  committed
  f229d2c

  Copy full SHA for f229d2c

• simplify profile text output by displaying only top-level ops statistics (#42262)

  Show description for 69080e9
  

  YazhiGao

  authored and

  facebook-github-bot

  committed
  69080e9

  Copy full SHA for 69080e9

• Update determinism documentation (#41692)

  Show description for d7ee84c
  

  kurtamohler

  authored and

  facebook-github-bot

  committed
  d7ee84c

  Copy full SHA for d7ee84c

• Fix exception chaining in torch/ (#43836)

  Show description for f17d7a5
  

  akihironitta

  authored and

  facebook-github-bot

  committed
  f17d7a5

  Copy full SHA for f17d7a5

• Move type annotations for remaining torch.utils stub files inline (#43406)

  Show description for da32bf4
  

  rgommers

  authored and

  facebook-github-bot

  committed
  da32bf4

  Copy full SHA for da32bf4

• [quant][graphmode][fix] Fix insert quant dequant for observers without qparams (#43606)

  Show description for 6022097
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  6022097

  Copy full SHA for 6022097

• [reland][quant][graphmode][fx] Add top level APIs (#43581) (#43901)

  Show description for 7db7da7
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  7db7da7

  Copy full SHA for 7db7da7

• [ONNX] Floordiv (#43022)

  Show description for ee53a33
  

  KsenijaS

  authored and

  facebook-github-bot

  committed
  ee53a33

  Copy full SHA for ee53a33

• Avoid resizing in MinMaxObserver (#43789)

  Show description for f73ba88
  

  Qi Zhou

  authored and

  facebook-github-bot

  committed
  f73ba88

  Copy full SHA for f73ba88

• [JIT] Remove loop peeling from the profiling executor pipeline. (#43847)

  Show description for 98b846c
  

  Mikhail Zolotukhin

  authored and

  facebook-github-bot

  committed
  98b846c

  Copy full SHA for 98b846c

• [JIT] Specialize autograd zero: actually remove the original graph after we created its versioned copy. (#43900)

  Show description for d69d603
  

  Mikhail Zolotukhin

  authored and

  facebook-github-bot

  committed
  d69d603

  Copy full SHA for d69d603

• [JIT] Specialize autograd zero: fix the guarding condition. (#43846)

  Show description for f150f92
  

  Mikhail Zolotukhin

  authored and

  facebook-github-bot

  committed
  f150f92

  Copy full SHA for f150f92

Commits on Aug 31, 2020

• Move torch/autograd/grad_mode.pyi stubs inline (#43415)

  Show description for 4c19a1e
  

  rgommers

  authored and

  facebook-github-bot

  committed
  4c19a1e

  Copy full SHA for 4c19a1e

• Implementing NumPy-like function torch.heaviside() (#42523)

  Show description for 3682df7
  

  Kiyosora

  authored and

  facebook-github-bot

  committed
  3682df7

  Copy full SHA for 3682df7

Pagination

Previous

Next