strisunshinewentingwang/pytorch

Pull Requests Commits

Extract TensorPipeAgent's collectNames to be a standalone utility function (#53202)

mrshenli committed 3 years ago

affdcce8

Automated submodule update: FBGEMM (#53478)

facebook-github-bot committed 3 years ago

e08aae26

[PyTorch Mobile] Allow skipping operator exists check when bytecode model is loaded (#52814)

dhruvbird committed 3 years ago

b26c0bb2

[Static Runtime] Call native resize_/resize_as_ as much as possible (#53425)

Hao Lu committed 3 years ago

3236efa4

[DataLoader] Introduce deterministic context (#53271)

ejguan committed 3 years ago

dbbe0a21

[DataLoader] ConcatDataPipe (#53301)

ejguan committed 3 years ago

1ba80264

[WIP][FX] Fix tracing support for torchbind (#52884)

James Reed committed 3 years ago

1fe6a650

Replace `internal::GRAIN_SIZE` by `grain_size` (parameter). (#53177)

ysiraichi committed 3 years ago

a0d1e701

[caffe2] Use extended versions of cuDNN calls for SpatialBN

valentinandrei committed 3 years ago

36960135

Prefix assert_async with underscore, fix some bugs in assert_async CUDA testing (#53276)

ezyang committed 3 years ago

758fb94f

Forbid trailing whitespace (#53406)

samestep committed 3 years ago

8c798e06

Revert D26849826: [pytorch][PR] Call nvidia-smi.exe before running tests Windows

ngimel committed 3 years ago

cab2689e

Cleanup async execution for python RPC calls. (#53230)

pritamdamania committed 3 years ago

19749698

Simplify async execution for script remote calls. (#53207)

pritamdamania committed 3 years ago

7bfa9dc7

Modify assert order to correct the error message when nan appears in multinomial on cuda (#53288)

xiaomengy committed 3 years ago

6cbbef2f

[caffe2] move the SaveOp implementation from a header to a .cc file (#53298)

simpkins committed 3 years ago

f595ba1b

docker: Update default cuda => 11.1 (#53299)

seemethere committed 3 years ago

474fe7d9

add distributed backend options in setup.py (#53214)

Rong Rong (AI Infra) committed 3 years ago

f58f7b78

Simplify async execution for script calls. (#53204)

pritamdamania committed 3 years ago

387d9a6b

automate sharding using S3 test time stats (#53269)

janeyx99 committed 3 years ago

c0adabe1

[caffe2] Fix shape inference for LpNorm (#53332)

khabinov committed 3 years ago

00bd0e98

Call nvidia-smi.exe before running tests Windows (#53334)

skyline75489 committed 3 years ago

efebc652

Revert "Automated submodule update: tensorpipe (#53012)" (#53394)

seemethere committed 3 years ago

c3405e5b

[1/n][torch/elastic][upstream] Move torchelastic/rendezvous to torch/distributed/rendezvous (#53172)

Kiuk Chung committed 3 years ago

ba75cedf

[DDP Logging] Log comm. hook in ddp logging (#52966)

rohan-varma committed 3 years ago

14fa4763

[DDP Logging] Log nccl_async_error_handling (#52965)

rohan-varma committed 3 years ago

5d9b7bee

[Dist Debugality] Log key DDP metrics to stderr under debug mode. (#52957)

rohan-varma committed 3 years ago

bdbfc258

Refactor/fix DDP model check during init (#52887)

rohan-varma committed 3 years ago

68134374

Properly skip distributed tests when distributed module is not built (#52945)

xuhdev committed 3 years ago

1b35b1a0

Refactor ForeachUnaryOp.cu (#51894)

izdeby committed 3 years ago

c697e480

Newer Older