strisunshinewentingwang/pytorch

Pull Requests Commits

[CI] Add TorchTrition conda packages (#89841)

malfet committed 1 year ago

59a2fe74

[Caffe2] Fix merge logic bug (#89551)

scramsby committed 1 year ago

24b3b73c

Remove beauby and dzdang from CODEOWNERS (#89811)

kit1980 committed 1 year ago

55789b40

[inductor] Add aten._native_batch_norm_legit to decomposition (#89843)

desertfire committed 1 year ago

693135a9

Update code style for optimizer code (#89862)

mlazos committed 1 year ago

3d47c74c

[quant][docs] Move some of the descriptions out of codeblock (#89795)

jerryzh168 committed 1 year ago

8ca09dda

Enable instance norm running mean test (#89793)

janeyx99 committed 1 year ago

fcb5d6e7

[FSDP] Another fix for `DTensor`, `use_orig_params=True` (#89845)

awgu committed 1 year ago

c599cf24

replace double transpose with single permute in nn.f.mha (#89847)

drisspg committed 1 year ago

b9afa928

Stream actually overrides __new__ so we need to patch it as well (#89592)

albanD committed 1 year ago

8713119c

Move gpu slow tests to sm86 (#87880)

davidberard98 committed 1 year ago

a029ec2c

Deprecating DataPipes (#89794)

ejguan committed 1 year ago

991028cd

Don't unsafely clone autograd meta (#89720)

ezyang committed 1 year ago

6c1fb3f2

Fix CopySlices logic to ensure wrapped node runs properly. (#89812)

albanD committed 1 year ago

02e2eaa9

[test_nn] split multihead_attention from test_nn (#89748)

kshitij12345 committed 1 year ago

8314d403

[Quant][docs] Use get_default_qconfig_mapping (#87299)

andrewor14 committed 1 year ago

fb47a669

[Quant][fx][bc-breaking] Remove backend_config_utils.py (#89810)

andrewor14 committed 1 year ago

2bce6d09

Revert "[GHA] Decrease Windows test timeout to 120 minutes (#89694)"

pytorchmergebot committed 1 year ago

e1dbd9a2

[FSDP] Relax post-backward assert (#89791)

awgu committed 1 year ago

6e2da426

Revert "Move functorch/_src to torch/_functorch (#88756)"

pytorchmergebot committed 1 year ago

218d9c6e

[follow-up] Python Attr Serialization (#88913)

kshitij12345 committed 1 year ago

086b251f

don't run input mutation analysis in dynamo (#89760)

bdhirsh committed 1 year ago

2f9ec226

[aarch64] add SLEEF dependency for aten_cpu (#89475)

ajaymh committed 1 year ago

3cef87f9

[Quant][docs] Fix BackendConfig example in docstring/README (#89319)

andrewor14 committed 1 year ago

c6ede0bd

Move functorch/_src to torch/_functorch (#88756)

zou3519 committed 1 year ago

52bc5c1c

Guard the boundary of index computed in compute_source_index_and_lambda (#89252)

jgong5 committed 1 year ago

620994cd

[PyTorch Edge] Set training for module only (#89488)

cccclai committed 1 year ago

93772305

Refactoring to share vectorization code for int8/uint8. (#89650)

alexsamardzic committed 1 year ago

a78467f3

[minifier] Continue on assertion for accuracy minification (#89739)

anijain2305 committed 1 year ago

8226a5d3

[dynamo] Don't copy the graph during checkpointing (copy_graphstate) (#89232)

mlazos committed 1 year ago

40dd03ee

Newer Older