Commits

Branch selector

 d00de0d

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 28, 2022

• Support dataclasses in TorchScript (pytorch#73066)

  Show description for d00de0d
  

  norabelrose

  authored and

  pytorchmergebot

  committed
  d00de0d

  Copy full SHA for d00de0d

• [ONNX] Capture annotated attributes for local function

  Show description for b3cfc74
  

  BowenBao

  authored and

  pytorchmergebot

  committed
  b3cfc74

  Copy full SHA for b3cfc74

• [Static Runtime] Add test that runs NNC fused kernels in parallel (pytorch#73256)

  Show description for cfd92f2
  

  navahgar

  authored and

  pytorchmergebot

  committed
  cfd92f2

  Copy full SHA for cfd92f2

• Add api for recursively analyzing function calls (pytorch#73329)

  Show description for ab6395f
  

  Elias Ellison

  authored and

  pytorchmergebot

  committed
  ab6395f

  Copy full SHA for ab6395f

• Allow custom registration of shape functions (pytorch#73270)

  Show description for d3d74e9
  

  Elias Ellison

  authored and

  pytorchmergebot

  committed
  d3d74e9

  Copy full SHA for d3d74e9

• Enable win-arm64

  Show description for c4ff49f
  

  gaborkertesz-linaro

  authored and

  pytorchmergebot

  committed
  c4ff49f

  Copy full SHA for c4ff49f

• [BE][GHA] Use timeout_after for win templates

  Show description for a1d5b5d
  

  malfet

  authored and

  pytorchmergebot

  committed
  a1d5b5d

  Copy full SHA for a1d5b5d

• fused int8 static (pytorch#73452)

  Show description for 1436507
  

  xiaosun86

  authored and

  pytorchmergebot

  committed
  1436507

  Copy full SHA for 1436507

• Add pickling support for WorkerInfo (pytorch#73371)

  Show description for 6c8e516
  

  H-Huang

  authored and

  pytorchmergebot

  committed
  6c8e516

  Copy full SHA for 6c8e516

• Remove legacy tensor constructors for complex dtypes

  Show description for f437ca6
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  f437ca6

  Copy full SHA for f437ca6

• Refactor set_default_tensor_type to avoid legacy tensor types

  Show description for d51103e
  

  peterbell10

  authored and

  pytorchmergebot

  committed
  d51103e

  Copy full SHA for d51103e

• Also install c10d headers with .h extension (pytorch#73422)

  Show description for b213041
  

  lw

  authored and

  pytorchmergebot

  committed
  b213041

  Copy full SHA for b213041

• [Static Runtime] Add auto-generated out variant dispatchers (pytorch#72603)

  Show description for fe7e1bd
  

  d1jang

  authored and

  pytorchmergebot

  committed
  fe7e1bd

  Copy full SHA for fe7e1bd

• [GHA] Increase win test timeout to 5h (pytorch#73490)

  Show description for 94501ff
  malfet

  authored
  94501ff

  Copy full SHA for 94501ff

• Update XNNPACK (pytorch#73467)

  Show description for a22d079
  

  malfet

  authored and

  pytorchmergebot

  committed
  a22d079

  Copy full SHA for a22d079

• Fix test_Sparse_to_Sparse_copy__cuda_bfloat16 failure (pytorch#73157)

  Show description for 3c932c3
  

  pearu

  authored and

  pytorchmergebot

  committed
  3c932c3

  Copy full SHA for 3c932c3

Commits on Feb 27, 2022

• [FSDP] full_state_dict impl (pytorch#73324)

  Show description for 540361f
  

  rohan-varma

  authored and

  pytorchmergebot

  committed
  540361f

  Copy full SHA for 540361f

• [FSDP] Generic arguments for state_dict (pytorch#73323)

  Show description for 7aec6d7
  

  rohan-varma

  authored and

  pytorchmergebot

  committed
  7aec6d7

  Copy full SHA for 7aec6d7

• [test] Disable TestXNNPACK on ROCM machines (pytorch#73476)

  Show description for 09697c6
  

  digantdesai

  authored and

  pytorchmergebot

  committed
  09697c6

  Copy full SHA for 09697c6

Commits on Feb 26, 2022

• [quant][fx] Add lowering support for functional ops using DefaultNodeQuantizeHandler (pytorch#73120)

  Show description for 5613527
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  5613527

  Copy full SHA for 5613527

• [fx/graph_drawer] Add args/kwargs and users (pytorch#73464)

  Show description for b196e01
  

  jfix71

  authored and

  pytorchmergebot

  committed
  b196e01

  Copy full SHA for b196e01

• [quant][fx] Add root_node_getter in backend_config_dict (pytorch#73345)

  Show description for 45a0420
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  45a0420

  Copy full SHA for 45a0420

• T112685841: Use irange in PyTorch (pytorch#73378)

  Show description for 05c86c2
  

  Shuitao Fan

  authored and

  pytorchmergebot

  committed
  05c86c2

  Copy full SHA for 05c86c2

• Fix test_sparse_addmm_...float16 and test_sparse_matmul_...float16 test failures (pytorch#73155)

  Show description for 16cd685
  

  pearu

  authored and

  pytorchmergebot

  committed
  16cd685

  Copy full SHA for 16cd685

• Fix test missing target (pytorch#73415)

  Show description for 186ef8b
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  186ef8b

  Copy full SHA for 186ef8b

• Sparse CSR CPU: implement addmm(dense, sparse, sparse) -> dense (pytorch#73076)

  Show description for 0ba3498
  

  IvanYashchuk

  authored and

  pytorchmergebot

  committed
  0ba3498

  Copy full SHA for 0ba3498

• Fix CUDA error when multiplying sparse hybrid tensors with zero dense dimensions (pytorch#73428)

  Show description for 4c52264
  

  pearu

  authored and

  pytorchmergebot

  committed
  4c52264

  Copy full SHA for 4c52264

• [SR] Disable aten::where out variant (pytorch#73367)

  Show description for d398d4d
  

  Mike Iovine

  authored and

  pytorchmergebot

  committed
  d398d4d

  Copy full SHA for d398d4d

• [PyTorch Distributed] Consolidate NCCL_DESYNC_DEBUG and TORCH_DISTRIBUTED_DEBUG=INFO (pytorch#73257)

  Show description for 2aade49
  

  kwen2501

  authored and

  pytorchmergebot

  committed
  2aade49

  Copy full SHA for 2aade49

• [PT-D][Sharding] Make Partial separate file and enable padding for reshard when size not divisible by world_size (pytorch#73392)

  Show description for 11224db
  

  fduwjj

  authored and

  pytorchmergebot

  committed
  11224db

  Copy full SHA for 11224db

• Don't use vector accessor methods to do pointer math; unblock platform010 (pytorch#73414)

  Show description for 33ca944
  

  r-barnes

  authored and

  pytorchmergebot

  committed
  33ca944

  Copy full SHA for 33ca944

• [Easy][c10d] Minor fixes (pytorch#73318)

  Show description for 12890ab
  

  awgu

  authored and

  pytorchmergebot

  committed
  12890ab

  Copy full SHA for 12890ab

Commits on Feb 25, 2022

• ci: Bump CUDA 11.1 -> 11.3

  Show description for 9a8000f
  

  seemethere

  authored and

  pytorchmergebot

  committed
  9a8000f

  Copy full SHA for 9a8000f

• [qunat][fx][fix] Fix get_module_type for fusion (pytorch#72735)

  Show description for 16554be
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  16554be

  Copy full SHA for 16554be

• Revert D33992795: Add BUILD_LAZY_CUDA_LINALG option

  Show description for 3127128
  

  janeyx99

  authored and

  pytorchmergebot

  committed
  3127128

  Copy full SHA for 3127128

Pagination

Previous

Next