strisunshinewentingwang/pytorch

Pull Requests Commits

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

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

[BackwardCompatibility] Remove aten::to from allow_list (#60147)

Hao Lu committed 3 years ago

2b51a8a9

[special] erfcx cuda support (#60519)

kshitij12345 committed 3 years ago

3ca28656

cuda rpc backward sparse tensor fix (#59609)

gcramer23 committed 3 years ago

46d27a53

Revert D29330585: [pytorch][PR] add BFloat16 support for arange on CPU

Mike Ruberry committed 3 years ago

561132f9

Introduce quantized convolution serialization format 3 (#60241)

dreiss committed 3 years ago

d63c236f

Newer Older