strisunshinewentingwang/pytorch

Pull Requests Commits

Fix missing #pragma once in jit/method.h

wconstab committed 3 years ago

367aff91

Add CUDA Vital (#58059)

Victor Bittorf committed 3 years ago

8b6487c6

add a boxed CPU fallback kernel (#58065)

bdhirsh committed 3 years ago

9134b0e4

[torch] Module fix on the support of LazyModule on bug #60132 (#60517)

Hongbo Zhang committed 3 years ago

ad69e2fd

faster generate_square_subsequent_mask in nn.Transformer (#60631)

Basil Hosmer committed 3 years ago

cab926b2

Remove `Optional[None]` annotations (#60704)

Ansley Ussery committed 3 years ago

7585783b

Fix doc preview source directory (#60792)

driazati committed 3 years ago

5ed7400b

Merge remote-tracking branch 'upstream/master' into ci-all/batchnorm3d-channels-last-cudnn-spatial-persistent

xwang233 committed 3 years ago

ad3a895f

configurable pre/post LayerNorm in nn.Transformer (#60593)

Basil Hosmer committed 3 years ago

7b933cd9

Revert "Revert D29135358: [quant] Input-Weight Equaliaztion - convert modifications" (#60646)

angelayi committed 3 years ago

e13a9587

[Tools] Update path in clang_format_utils after #60473 (#60782)

Mikhail Zolotukhin committed 3 years ago

7188d84c

[caffe2] update make_cifar_db to move the string into DB::Put() (#60692)

simpkins committed 3 years ago

394f60b0

To intorduce Functional API for multi-tensor (#60735)

iramazanli committed 3 years ago

e1bd4963

Add missing kernel checks (#60635)

r-barnes committed 3 years ago

8f16a380

Faster cumsum and cumprod backwards (#60642)

lezcano committed 3 years ago

dfc8247d

Use S3 for documentation previews (#60711)

driazati committed 3 years ago

d3bec9f4

Dispatch to Python via __torch_dispatch__ (#59760)

ezyang committed 3 years ago

aacc722a

Remove test linalg test skips from MAGMA integration (#58232)

amathews-amd committed 3 years ago

a53d7f8f

Use python3.6 compatible APIs in clang_tidy.py (#60659)

1ntEgr8 committed 3 years ago

8216da1f

Add python version and cuda-specific folder to store extensions (#60592)

andfoy committed 3 years ago

6322f668

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

crcrpar committed 3 years ago

a404cc9a

[c10d] Fix test_collective_hang flakiness (#60662)

rohan-varma committed 3 years ago

0be65cd5

Add --print-include-paths option to tools/linter/clang_tidy.py (#60744)

1ntEgr8 committed 3 years ago

474bdaf5

Fix --dry-run option in tools/linter/clang_tidy.py (#60744)

1ntEgr8 committed 3 years ago

608f12b8

Parametrizations depending on several inputs (#60530)

lezcano committed 3 years ago

3a838e4c

Fix incorrect doc about the dtype for `torch.randint` described in issue #56347 (#60507)

NivekT committed 3 years ago

8cba3653

[Delegate] Support composite of lowered sub modules of the same backend (#59921)

iseeyuan committed 3 years ago

d8c3d555

To refactor Sparse Adam algorithm for functional form (#59171)

iramazanli committed 3 years ago

7c2938bf

Improve numerical stability of LayerNorm (#59987)

xiaomengy committed 3 years ago

963c9833

When creating a single parition skip the output nodes, but process possible nodes after it. (#60370)

Protonu Basu committed 3 years ago

5b1f5c8f

Newer Older