strisunshinewentingwang/pytorch

Pull Requests Commits

Add AST rewriter to acc_tracer (#53644)

jfix71 committed 3 years ago

56f3cb7a

[JIT] [Remove Mutation] Add handling of normal_ (#52175)

Elias Ellison committed 3 years ago

5563248b

update upsample tests in test_nn.py to test for memory_format (#53665)

bdhirsh committed 3 years ago

c68cc24c

Replace supports_tensor_out with supports_out (#53745)

imaginary-person committed 3 years ago

669fcf30

Fix distributions which don't properly honor validate_args=False (#53600)

t-vi committed 3 years ago

76b58dd9

[2/n][torch/elastic][upstream] Move torchelastic/timer torchelastic/multiprocessing to torch/distributed/elastic (#53574)

Kiuk Chung committed 3 years ago

b03c92a9

Port NumPy typing testing style to PyTorch (#52408)

guilhermeleobas committed 3 years ago

cb680393

[package] make importer a little more obscure (#51676)

suo committed 3 years ago

17bc70e6

[package] implement `get_resource_reader` API (#51674)

suo committed 3 years ago

b4d8f4af

Give line numbers in git-grep-based lints (#53733)

samestep committed 3 years ago

bfc80b35

Simplify init._calculate_fan_in_and_fan_out (#53522)

JeanKossaifi committed 3 years ago

70a43425

clang format reducer and logger files (#53148)

zhaojuanmao committed 3 years ago

a76b4736

fix data type logging (#53162)

zhaojuanmao committed 3 years ago

d032287e

add is_multi_device_module logging field (#53149)

zhaojuanmao committed 3 years ago

7d4b229d

allow users to set sample rate and add per iteration latency breakdowns (#53145)

zhaojuanmao committed 3 years ago

a08fc1a7

eig_backward: faster and with complex support (#52875)

nikitaved committed 3 years ago

8f15a2f0

Removes trailing whitespace (#53728)

Mike Ruberry committed 3 years ago

b99b6065

[andoid] publishing to maven central (#53568)

IvanKobzarev committed 3 years ago

5658ab5f

[android] bump gradle version to 6.8.3 (#53567)

IvanKobzarev committed 3 years ago

05e0ea96

Use native ctc loss for target length 256 (#53557)

t-vi committed 3 years ago

e13ef777

Added CUDA support for torch.orgqr (#51348)

IvanYashchuk committed 3 years ago

e937db5d

[fix] nn.Embedding: allow changing the padding vector (#53447)

kshitij12345 committed 3 years ago

45ddf113

Improve logic for S3 stats gathering. Uses automatic SLOW_TESTS. (#53549)

janeyx99 committed 3 years ago

bcbe0720

Remove reference to 9.2 as it's been removed from nightlies (#53716)

janeyx99 committed 3 years ago

1c9fc38e

Marginally improve pytest collection for top-level test files (#53617)

ezyang committed 3 years ago

70733f2e

Fix inaccurate dispatch table for fill_ (#53611)

ezyang committed 3 years ago

6e020a48

Convert a few more checks to raise NotImplementedError (#53610)

ezyang committed 3 years ago

4dbd0b63

[RELAND] Deduplicate shared params before constructing Reducer in DDP (#53279)

mcarilli committed 3 years ago

e787872a

If distributed module isn't available, don't run distributed/pipeline tests (#53547)

xuhdev committed 3 years ago

039402b9

Filter 0's returned by exponential distribution (#53480)

ngimel committed 3 years ago

6aa5148d

Newer Older