strisunshinewentingwang/pytorch

Pull Requests Commits

TransformedDistribution.icdf: Fix erroneous icdf ValueError (#71393)

pejato committed 2 years ago

8548657d

[FSDP] Backward prefetch in recursive call (#71804)

rohan-varma committed 2 years ago

d0ff1f00

[FSDP] Add/refactor unit test for wrap (#71803)

rohan-varma committed 2 years ago

a30b0cf5

[Easy] Format DDP error (#71802)

rohan-varma committed 2 years ago

99a99292

Revert D33805315: [pytorch][PR] Automated submodule update: FBGEMM

malfet committed 2 years ago

f115a423

Fix forward AD for cudnn batch norm (#71901)

soulitzer committed 2 years ago

51ae9ccb

[GHF] More verbose failures messages (#71941)

malfet committed 2 years ago

3b9f2e2c

Fix bug where frozen mod not used for OFI #68903 (#71436)

Gamrix committed 2 years ago

c8596560

[jit][edge] Make flatbuffer_serailzer print correct type strings. (#71935)

zhxchen17 committed 2 years ago

b4867978

Remove unnecessary non_contiguous and gradient tests from test_linalg (#68188)

lezcano committed 2 years ago

1407939f

Generalize noncontiguous tests to several outputs (#67996)

lezcano committed 2 years ago

6cb128c8

Make repeat_interleave respect the conj and neg bits. (#68523)

lezcano committed 2 years ago

171cf153

Deactivate the tracking of gradients in sampling functions within OpInfos (#68522)

lezcano committed 2 years ago

a675770a

Add OpInfo test to check that floating point inputs in OpInfos have requires_grad set to True (#69909)

lezcano committed 2 years ago

e2011b29

Implement derivatives for torch.remainder and torch.fmod wrt the second argument and update the docs (#69908)

lezcano committed 2 years ago

dcc6aed5

[warnings] Disable broken TORCH_CHECK (#71947)

Nolan O'Brien committed 2 years ago

b62780fc

Automated submodule update: FBGEMM (#65595)

facebook-github-bot committed 2 years ago

75cc2184

Add complex support for Jiterator, port sinc to Jiterator (#71577)

zasdfgbnm committed 2 years ago

81d1ce05

[c10d] Enable gather_object on nccl (#71623)

wanchaol committed 2 years ago

8551989b

Try setting pull_request checkout to head ref (#71734)

suo committed 2 years ago

e27271d0

Update the operator version check logic when generating models for testing upgraders (#71894)

cccclai committed 2 years ago

e755a4f1

Improved error messages for `max_unpool{}d` operators (#67328)

krshrimali committed 2 years ago

0cae3c04

Added antialias flag to interpolate (CUDA, bilinear and bicubic) (#70930)

vfdev-5 committed 2 years ago

eeda31fa

Support printing inplace operators in FX (#71887)

jansel committed 2 years ago

567c2bb8

Revert D33820822: [pytorch][PR] Run the pr-label check on PR closed action and validate closed_by

atalman committed 2 years ago

5a14eca1

Implement scatter primitive for ProcessGroupNCCL (#70029)

wanchaol committed 2 years ago

6feba4bc

Implement gather primitive for ProcessGroupNCCL (#66745)

wanchaol committed 2 years ago

9b53d319

ci: Enable tests for iOS on GHA

seemethere committed 2 years ago

0a8b3919

Run the pr-label check on PR closed action and validate closed_by (#71917)

atalman committed 2 years ago

e020414c

Implement forward AD for linalg.svd and improve svd_backward (#70253)

lezcano committed 2 years ago

8ff1a8fd

Newer Older