strisunshinewentingwang/pytorch

Pull Requests Commits

Support NCCL Premul Sum (#81272)

crcrpar committed 2 years ago

432c508e

Support the XPU backend untyped storage (#83952)

chengjunlu committed 2 years ago

67aed393

Support returning symbolic strides from t.stride() in Python (#83842)

ezyang committed 2 years ago

0491e1a1

[BE][CUDA] Use packed_accessor64 (#83949)

malfet committed 2 years ago

df70714e

Remove conj kernels for real dtypes (#80374)

peterbell10 committed 2 years ago

754d7f05

[Nested Tensor] Make offset copy and move assignment more explicit. (#83488)

drisspg committed 2 years ago

2c76d05b

Task: T129772171 remove assertEqualIgnoreTypes from test/test_nn.py (#83870)

Ishan-Rajgarhia committed 2 years ago

7fdc2f70

Move nnapi code from ATen common code to specific library (#83748)

kirklandsign committed 2 years ago

6edcf8e1

add merge blocking to ci: sev template (#83940)

clee2000 committed 2 years ago

84f0411f

[fbia] Keep Track of full qualified name before and after remote sharding (#83889)

nanoax committed 2 years ago

c47e0450

Add hypothesis to requirements.txt (#83740)

ezyang committed 2 years ago

58f61d50

Revert "Optimize transpose copy on CPU using fbgemm transpose (#83327)"

pytorchmergebot committed 2 years ago

84e45e7e

Fix use-dict-literal lint (#83718)

kit1980 committed 2 years ago

591222f5

Back out "Support regex-style matching for Any and Oneof (#82853)" (#83922)

Shirong Wu committed 2 years ago

fc470cf9

[fx][pass infra] Adding error catching (#83933)

angelayi committed 2 years ago

89072177

ci: Remove dead code related to android uploads (#83930)

seemethere committed 2 years ago

7c8d2658

Remove CoreMLMemoryObserver (#83703)

jmdetloff committed 2 years ago

21bc77ca

Don't introduce new overload for SymInt (#83628)

ezyang committed 2 years ago

8fae7027

Prefer signal from land checks over PR signals (#83715)

ZainRizvi committed 2 years ago

4808bda7

Fix load_extra_only api for flatbuffers and enable flatbuffers in mobile for OSS properly (#83855)

chenlai committed 2 years ago

25dd2a04

Revert "Strenghten preconditions of linalg.cross (#83798)"

pytorchmergebot committed 2 years ago

bbe803cb

[complex] conv_transpose1d (#79694)

kshitij12345 committed 2 years ago

a802603e

[fix] edge case in `MaxPool1d` and add ErrorInputs (#83553)

khushi-411 committed 2 years ago

90950302

remove assertEqualIgnoreTypes from test/distributions/test_distributions.py (#83709)

liuk22 committed 2 years ago

8f9ae356

[torchgen] Add documentation for `autogen` keyword (#83610)

larryliu0820 committed 2 years ago

5204b8e4

[vulkan] Add VMA as a third_party subrepo (#83906)

SS-JIA committed 2 years ago

732255f0

Fix view_func replay in no-grad mode (#83872)

soulitzer committed 2 years ago

81843596

Strenghten preconditions of linalg.cross (#83798)

lezcano committed 2 years ago

7f0198e7

Enable NCCL_DESYNC_DEBUG when TORCH_DISTRIBUTED_DEBUG=DETAIL (#83881)

kwen2501 committed 2 years ago

9beddde1

Allow None arguments for elementwise type promotion wrapper and fix clamp with None arguments (#83586)

IvanYashchuk committed 2 years ago

cb488e6d

Newer Older