strisunshinewentingwang/pytorch

Pull Requests Commits

Fix the slowdown of _object_to_tensor since 1.9 (#65721)

Yi Wang committed 2 years ago

12137db5

[acc_utils] Fix off by one for model info getter (#65708)

jfix71 committed 2 years ago

002ff198

Refactor AotCompile to return a pair (#65707)

priyaramani committed 2 years ago

63bb7c6d

Add nn.function.hardtanh in acc_tracer (#65639)

zrphercule2 committed 2 years ago

e9327ed2

[quant] Add op benchmark for CPU FakeQuantizePerChannel with float zero_points (#65241)

b-koopman committed 2 years ago

6a6ee92e

add deepcopy support to subclasses (#65584)

albanD committed 2 years ago

7c62b6e9

Sparse SoftMax: Remove unused variables (#65539)

peterbell10 committed 2 years ago

f5b4e369

[GHA] Small refactors (#65647)

malfet committed 2 years ago

e1340d42

Add HPU type for check_base_legacy_new (#65410)

sujoysaraswati committed 2 years ago

fea32be9

Apply linter suggestions to #65137 (#65459)

malfet committed 2 years ago

82e0bf44

Upload sccache stats (#65582)

driazati committed 2 years ago

811601e1

[Reland] nn.functional.linear OpInfo (#65498)

zou3519 committed 2 years ago

ea546e20

upgrade windows cuda installer: cu11.1.0 to cu11.1.1 (#65669)

mszhanyi committed 2 years ago

b91375f7

[package] add some docs describing how to debug dependencies (#65704)

suo committed 2 years ago

cd2656a2

Avoid storage access for HPU tensors (#65409)

sujoysaraswati committed 2 years ago

10d0dbc6

Remove confusing SHARD_NUMBER resetting logic (#65701)

janeyx99 committed 2 years ago

aa5d2a8d

Update ProcessGroup collective C++ APIs to be non-pure virtual functions (#64943)

mrshenli committed 2 years ago

facff2ec

Bug fixes in dataframe_wrapper (#65629)

nayef211 committed 2 years ago

cd80bbe5

[BE] Run Zero test internally (#65519)

rohan-varma committed 2 years ago

1c8949c5

[BE] Enable ZeRO test on windows (#65385)

rohan-varma committed 2 years ago

f70147b4

[Codemod][FBSourceBlackLinter] Daily `arc lint --take BLACK`

CodemodService Bot committed 2 years ago

4fe66d96

Add all_paths utility function (#65602)

tugsbayasgalan committed 2 years ago

146817c9

[TensorExpr] Delete dtype_ field from Let - it should use its var's dtype. (#65634)

Mikhail Zolotukhin committed 2 years ago

0256c3be

Revert D31172530: [pytorch][PR] Enable CUPTI for kineto by default on windows

malfet committed 2 years ago

399214ef

Add nn.function.hardswish in acc_tracer (#65590)

zrphercule2 committed 2 years ago

cda2ee90

Add quantized::convtranspose2d (#63914)

Akshit Khurana committed 2 years ago

1de8976e

add qmul (#63913)

Akshit Khurana committed 2 years ago

ab5eb569

[PyTorch] Store Argument::alias_info_ on the heap (#64824)

swolchok committed 2 years ago

ece25c45

Rocm4.3.1 nightly (#65624)

KyleCZH committed 2 years ago

af7238f2

[TensorExpr] Re-enable a float16 test. (#65632)

Mikhail Zolotukhin committed 2 years ago

15724bcc

Newer Older