Commits

Branch selector

 master

User selector

crcrpar

Datepicker

All time

Commit History

Commits on Oct 3, 2022

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

  Show description for 85d520d
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  85d520d

  Copy full SHA for 85d520d

Commits on Sep 30, 2022

• Remove `optim_mt` from `test/test_optim.py` (#83549)

  Show description for 4c04fa9
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  4c04fa9

  Copy full SHA for 4c04fa9

Commits on Sep 29, 2022

• Use `int64_t` for nll_loss with cuda inputs (#85395)

  Show description for ef0baba
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  ef0baba

  Copy full SHA for ef0baba

• resubmit: "resubmit: [mta] APEX style Fused Adam (#81705) (#85507)" (#85739)

  Show description for 5f26df0
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  5f26df0

  Copy full SHA for 5f26df0

Commits on Sep 23, 2022

• resubmit: [mta] APEX style Fused Adam (#81705) (#85507)

  Show description for 4615d1b
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  4615d1b

  Copy full SHA for 4615d1b

Commits on Sep 20, 2022

• [mta] APEX style Fused Adam (#81705)

  Show description for 7a6c4d0
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  7a6c4d0

  Copy full SHA for 7a6c4d0

Commits on Sep 14, 2022

• `ReflectionPad` supports `BFloat16` (#84949)

  Show description for 26b5986
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  26b5986

  Copy full SHA for 26b5986

Commits on Sep 2, 2022

• Add NCCL PreMul Sum to c10d `redce` ops (#84243)

  Show description for ab6c572
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  ab6c572

  Copy full SHA for ab6c572

Commits on Aug 24, 2022

• Support NCCL Premul Sum (#81272)

  Show description for 432c508
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  432c508

  Copy full SHA for 432c508

Commits on Aug 3, 2022

• [foreach][mta] Inplace `maximum` and `minimum` (#82523)

  Show description for 3139722
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  3139722

  Copy full SHA for 3139722

Commits on Jun 28, 2022

• `__launch_bounds__` for `torch.mode` with CUDA 11.7 (#79710)

  Show description for 7549f90
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  7549f90

  Copy full SHA for 7549f90

Commits on May 2, 2022

• [torch.onnx] support `torch.nn.functional.grid_sample`

  Show description for 0ae3aa6
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  0ae3aa6

  Copy full SHA for 0ae3aa6

Commits on Mar 23, 2022

• [Foreach Reduction] Use `OpMathType` tensor for intermediate results

  Show description for 105e58a
  

  crcrpar

  authored and

  pytorchmergebot

  committed
  105e58a

  Copy full SHA for 105e58a

Commits on Dec 17, 2021

• Fix build on latest main branch of thrust - SoftMax.cu (#70039)

  Show description for 1065739
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  1065739

  Copy full SHA for 1065739

Commits on Nov 19, 2021

• [Docs] Mention `torch.bfloat16` in `torch.finfo` (#68496)

  Show description for 9ce3c63
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  9ce3c63

  Copy full SHA for 9ce3c63

Commits on Nov 9, 2021

• Disable TF32 in `pinv_jvp` and `pinv_backward` (#67948)

  Show description for c5e5264
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  c5e5264

  Copy full SHA for c5e5264

Commits on Nov 5, 2021

• [Foreach] Implement L1&L2 norm (#62646)

  Show description for cdd5d16
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  cdd5d16

  Copy full SHA for cdd5d16

Commits on Nov 4, 2021

• Resubmit #67161 (#67735)

  Show description for 7c739e1
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  7c739e1

  Copy full SHA for 7c739e1

Commits on Oct 30, 2021

• make `TORCH_(CUDABLAS|CUSOLVER)_CHECK` usable in custom extensions (#67161)

  Show description for d4493b2
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  d4493b2

  Copy full SHA for d4493b2

Commits on Sep 30, 2021

• fix typo in _sharded_tensor (#65511)

  Show description for 768cfaa
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  768cfaa

  Copy full SHA for 768cfaa

Commits on Aug 18, 2021

• Use `fastAtomicAdd` in EmbeddingBag (mode "max") backward (#63298)

  Show description for 565578c
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  565578c

  Copy full SHA for 565578c

Commits on Aug 17, 2021

• `F.avg_pool3` CUDA backward: gpuAtomicAddNoReturn -> fastAtomicAdd (#63387)

  Show description for da87d64
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  da87d64

  Copy full SHA for da87d64

Commits on Jul 30, 2021

• [Foreach] support implicit broadcasting in slow path (#62167)

  Show description for 43d4fe6
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  43d4fe6

  Copy full SHA for 43d4fe6

Commits on Jul 22, 2021

• Foreach Test Refactor: Pointwise, Min/Max-imum (#61327)

  Show description for 8a2063e
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  8a2063e

  Copy full SHA for 8a2063e

Commits on Jul 12, 2021

• Add `expecttest` to CONTRIBUTING.md (#61163)

  Show description for 7136a62
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  7136a62

  Copy full SHA for 7136a62

Commits on Jul 6, 2021

• Foreach Binary Test Refactor (#59907)

  Show description for fac744e
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  fac744e

  Copy full SHA for fac744e

Commits on Jun 25, 2021

• CUDA `addcmul` and `addcdiv` do math in float for 16 bits I/O (#60715)

  Show description for a404cc9
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  a404cc9

  Copy full SHA for a404cc9

Commits on Jun 24, 2021

• Use `accscalar_t` for CUDA add/sub with Tensor and Scalar (#60454)

  Show description for 9e773ea
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  9e773ea

  Copy full SHA for 9e773ea

Commits on Jun 21, 2021

• [add/sub] Cast `alpha` to `acc_type` (#60227)

  Show description for b298013
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  b298013

  Copy full SHA for b298013

Commits on Jun 20, 2021

• Replace TensorRT's deprecated API in `caffe2/python/trt/test_pt_onnx_trt.py` (#60236)

  Show description for c19acf8
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  c19acf8

  Copy full SHA for c19acf8

Commits on Jun 9, 2021

• Refactor Foreach Tests: Unary Functions (#58960)

  Show description for 0897df1
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  0897df1

  Copy full SHA for 0897df1

Commits on May 25, 2021

• [PyTorch] Reduce errors of `foreach` functions (#56993)

  Show description for 7eade66
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  7eade66

  Copy full SHA for 7eade66

Commits on May 20, 2021

• Add `#pragma once` to CUDA foreach headers (#58209)

  Show description for ce3788d
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  ce3788d

  Copy full SHA for ce3788d

Commits on Nov 14, 2020

• Update links in DDP note (#47663)

  Show description for 2eb1e86
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  2eb1e86

  Copy full SHA for 2eb1e86

Commits on Aug 6, 2020

• Implement `gpu_kernel_multiple_outputs` (#37969)

  Show description for eb9ae7c
  

  crcrpar

  authored and

  facebook-github-bot

  committed
  eb9ae7c

  Copy full SHA for eb9ae7c

Pagination

Previous

Next