strisunshinewentingwang/pytorch

Pull Requests Commits

[xplat][pytorch]: Disabling too many logging. (#65170)

beback4u committed 3 years ago

cce53812

delegate parallelism to Ninja when possible (#64733)

Michael Dagitses committed 3 years ago

047e6823

add test for number of jobs when building (#65162)

Michael Dagitses committed 3 years ago

b936a100

Remove CUDA 9.2 references conditionals and workarounds (#65070)

janeyx99 committed 3 years ago

1ee66a52

fix torch.distributed.elastic event docs (#64974)

edward-io committed 3 years ago

51e12f00

[nnc] Updated inlining to handle cases when producer indices are constants after eval (#65044)

navahgar committed 3 years ago

bbe25af0

[nnc] Updated inliner to remove assertions and exception (#64719)

navahgar committed 3 years ago

03fc636d

[ONNX] Do not use `numpy` in ONNX opsets (#65188)

malfet committed 3 years ago

340531f2

[CoreML][OSS] Include Core ML in iOS/MacOS nightlies (#65075)

xta0 committed 3 years ago

7ced25ee

add a test case for const fold (#65224)

842974287 committed 3 years ago

f9c0a39a

[PyTorchEdge] promote prim ops by using ops table for mobile runtime (#64816)

pavithranrao committed 3 years ago

3c003aa6

Revert D30993855: [pytorch][PR] OpInfo: nn.functional.conv2d

suo committed 3 years ago

ecfc784e

[CoreML][OSS] Integrate with CMake (#64523)

xta0 committed 3 years ago

18fa58c4

[Reland] [Model Averaging] Simplify PostLocalSGD Optimizer API (#65197)

Yi Wang committed 3 years ago

c1415a0a

Bf16 matmul (#64619)

zhuhaozhe committed 3 years ago

752a8202

Torchhub: rewrite commit hash check to avoid using unnecessary GitHub API credits (#64362)

NicolasHug committed 3 years ago

f9bf144a

[FX] Ensure BC coverage for all of torch.fx.passes (#65081)

James Reed committed 3 years ago

0559cb37

[FX] Move graph_manipulation and param_fetch out of experimental and into passes (#65183)

James Reed committed 3 years ago

cf7409e1

[fx2trt] make gpu trace better (#65168)

842974287 committed 3 years ago

6aa04b68

[CoreML][iOS/MacOS] Add the CoreML executor (#64522)

xta0 committed 3 years ago

a8d7b885

Allow extra unused arguments in symbolic shape function (#65095)

Elias Ellison committed 3 years ago

aafeea3a

Actually deprecate __torch_function__ as plain methods (#64843)

albanD committed 3 years ago

6eafe7f1

Update fx proxy to use classmethod for __torch_function__ (#64842)

albanD committed 3 years ago

1ed9c33d

Use classmethods for overrides (#64841)

albanD committed 3 years ago

473e55d5

Fix port allocation race condition for elastic test (#65149)

H-Huang committed 3 years ago

a95fabfe

Small improvements to compare_models_torch binary (#65171)

SS-JIA committed 3 years ago

f1010705

Disable autograd fallback tests on Windows (#65147)

ezyang committed 3 years ago

9601deb1

implement "xy" indexing for torch.meshgrid (#62724)

Michael Dagitses committed 3 years ago

aaffcfe9

Allow parametrization to be nested (#65167)

albanD committed 3 years ago

d37c02be

Pass GITHUB_TOKEN to linux CI jobs and avoid skipping torchhub tests (#64807)

NicolasHug committed 3 years ago

9157a288

Newer Older