strisunshinewentingwang/pytorch

Pull Requests Commits

OpInfo for nn.functional.softmax (#62077)

krshrimali committed 3 years ago

09d10c43

[docs] Update sphinx to 3.5.4 (#61601)

garymm committed 3 years ago

9fdf7ec6

Clean up running smoke tests logic for Windows GHA (#62344)

janeyx99 committed 3 years ago

e352585f

Fix cppdoc example syntax (#62385)

chengscott committed 3 years ago

329426c2

[Linalg] Add cusolver syevjBatched path for torch.linalg.eigh when cuda >= 11.3 U1 (#62003)

xwang233 committed 3 years ago

d57ce8cf

[dte] fastpath implementations for mulgrad / divgrad (3/x) (#62437)

smacke committed 3 years ago

956c22b1

Remove an unused variable

lanza committed 3 years ago

607d720b

[quant] update per-channel observer min/max_val attribute names (#62345)

supriyar committed 3 years ago

cfd0f5eb

[sharded_tensor] add new init_from_local_shards API (#60479)

wanchaol committed 3 years ago

d92301dd

Fix setArgumentNames and make Script/Python consistent (#62442)

wconstab committed 3 years ago

bc787f24

[Prototype] [PyTorch Edge] Speed up model loading by 12% by directly calling the C file API from FileAdapter (#61997)

dhruvbird committed 3 years ago

725d98ba

[PyTorch Edge] Cache operator lambda during model loading [7% faster model loading] (#61996)

dhruvbird committed 3 years ago

693d8f2f

[PyTorch Edge] Add test for lite interpreter operator caching (#62306)

dhruvbird committed 3 years ago

0b3f42fa

[PyTorch Edge] Add test_lite_interpreter to fbsource xplat BUCK files (#62305)

dhruvbird committed 3 years ago

0bbdf0e1

Remove an unused variable

lanza committed 3 years ago

90977e10

[quant][graphmode][fx] Fix the calls to load_arg in quantization_patterns.py (#62376)

jerryzh168 committed 3 years ago

74291c83

[dte] broadcast fastpath implementations for reduce utility functions (2/x) (#62428)

smacke committed 3 years ago

eef85f89

[quant][graphmode] Add support for reference pattern for default ops (#62375)

jerryzh168 committed 3 years ago

21991770

[DDP Communication Hook] Simplify the implementation of parseHookResult of PythonCommHook (#62389)

Yi Wang committed 3 years ago

acba9b31

Reformat test_c10d_nccl.py and distributed_test.py (#62388)

Yi Wang committed 3 years ago

554daef8

[Model Averaging] Fix docstring of PeriodicModelAverager (#62392)

Yi Wang committed 3 years ago

9fee176b

[quant][graphmode] Add support for reference pattern for torch.cat (#62374)

jerryzh168 committed 3 years ago

8f519c5e

Change torch::Tensor to at::Tensor to fix build failure (#62425)

heitorschueroff committed 3 years ago

502823c2

Reland D29943356: .github: Migrate ecr_gc to github actions (#62438)

seemethere committed 3 years ago

49dc8277

[quant][graphmode] relax the constraint for supported_dtypes for reference option (Linear and Conv) (#62348)

jerryzh168 committed 3 years ago

dc8b5db5

[dte] scaffolding for c2 operator broadcasting fastpath (1/x) (#62369)

smacke committed 3 years ago

9f9244aa

Back out D29792193 "Add default Saved Variable hooks" (#62415)

yuguo68 committed 3 years ago

5c47038d

Back out D29848525 "Catch saved tensors default hooks race condition" (#62414)

yuguo68 committed 3 years ago

dcfcefcd

[reland] Refactor Tensor::to to call a primitive that is not copy_. (#62262)

zou3519 committed 3 years ago

389380ff

[jit] Renamed prim::Concat as prim::VarConcat (#61983)

navahgar committed 3 years ago

7b6d569a

Newer Older