strisunshinewentingwang/pytorch

Pull Requests Commits

[Profiler] Include ActivityType from Kineto (#80750)

Taylor Robie committed 2 years ago

2f6f7391

Don't implicitly convert to channels-first in MaxPool3D on CUDA (#80748)

eqy committed 2 years ago

3b78c568

[torchdynamo hash update] update the pinned torchdynamo hash (#81076)

pytorchmergebot committed 2 years ago

01a0dfb7

Disable functorch modes in testing's freeze_rng_state() (#81006)

zou3519 committed 2 years ago

26582056

Build example_allreduce only for GLOO (#81062)

kit1980 committed 2 years ago

b0aaefb5

reinstate size and shape returning symints (#79560)

Krovatkin committed 2 years ago

8389ccbc

[torchdynamo hash update] update the pinned torchdynamo hash (#80927)

pytorchmergebot committed 2 years ago

12632473

Run mask test with and without nested tensor (#81008)

mikekgfb committed 2 years ago

25449292

[MPS][BE] Introduce MPSUnaryCachedGraph (#81033)

malfet committed 2 years ago

e454be5f

torch.jit doc link for nvfuser readme.md (#77780)

jjsjann123 committed 2 years ago

d2c726d4

[cuDNN V8 API] (reopen 2) Allow the number of kernels profiled under torch.backends.cudnn.benchmark = True to be limitedCudnnv8 benchmark limit (#78299)

eqy committed 2 years ago

ae6dd20b

[Release Notes] Add way to export result from Google Sheets to Markdown (#79911)

Gamrix committed 2 years ago

7fd0cf55

[Release Notes] Simple script to merge categories (#79910)

Gamrix committed 2 years ago

6d4410b8

Adding additional topics to align with github topics list (#79909)

Gamrix committed 2 years ago

75586106

[Release Notes] Add Github PR link to csv export (#79908)

Gamrix committed 2 years ago

770fc74e

[Release Notes] Adding CSV Category Export (#78212)

Gamrix committed 2 years ago

ad6328ea

Always use the CommitCache, and make it a singleton (#78203)

Gamrix committed 2 years ago

62bf8071

Adding Author and Accepters information into pytorch release notes gen (#78190)

Gamrix committed 2 years ago

da549f58

Refactoring release not script to use dataclasses and have a shorter test. (#78189)

Gamrix committed 2 years ago

8549fafd

Update NestedTensor docs (#80963)

cpuhrsch committed 2 years ago

c97ff3d5

[onnx] Add argsort support (#80234)

shubhambhokare1 committed 2 years ago

712b3e76

[vulkan] Improve mutex management when syncing with GPU (#80959)

SS-JIA committed 2 years ago

31f98766

Fixing the torch.jit.freeze docs (#81020)

Gamrix committed 2 years ago

d2018f88

[Profiler] Add Pattern that detects extra cuda copy (#80572)

davidchencsl committed 2 years ago

39db8b38

[Profiler] Define pattern matcher structure (#80108)

davidchencsl committed 2 years ago

69a77f77

[composite compliance] as_strided (#80276)

kshitij12345 committed 2 years ago

4cd848ec

Setup for running PyTorch tests with TorchDynamo and skips for known failing tests (#80106)

anijain2305 committed 2 years ago

1d90d6ee

Update package.rst to not include hermetic claim (#81019)

PaliC committed 2 years ago

bd6bea35

nvfuser opinfo patch test_nvfuser_extremal_values_native_layer_norm_cuda (#80440)

jjsjann123 committed 2 years ago

e4eb95d4

Change cudnn incompatibility message wording (#80877)

atalman committed 2 years ago

a2ee1a92

Newer Older