strisunshinewentingwang/pytorch

Pull Requests Commits

Remove unnecessary header (#86212)

zou3519 committed 2 years ago

4f95f7ae

enable cpu meta testing (#86226)

albanD committed 2 years ago

6d7235e3

Add ref for cumsum (#86229)

lezcano committed 2 years ago

1432b997

Fix default correction value in std/var decompositions (#85839)

peterbell10 committed 2 years ago

b317736c

[AO] Cubic sparsity level scheduler (#85232)

z-a-f committed 2 years ago

adb12438

[FSDP] Expose internal prefetch limits (#86198)

awgu committed 2 years ago

24879698

Fix ITT unit-tests if PyTorch is compiled with `USE_ITT=OFF` (#86199)

jingxu10 committed 2 years ago

f20e4eab

[6/N] [Dispatchable Collectives] Update recv with CPU / CUDA implementations (#83876)

H-Huang committed 2 years ago

d39e9c1e

[kineto] make ProfilerKineto the only option (#84714)

chaekit committed 2 years ago

d447eff1

Adding Wunused-local-typedef build flag (#86154)

mehtanirav committed 2 years ago

d724a919

[MPS] Remove incorrect asserts from `Copy.mm` (#86184)

malfet committed 2 years ago

8da704cd

Add device logic handling for functions which allow scalar inputs as tensors (#86149)

eellison committed 2 years ago

9da5646c

[primTorch] special: j0, j1, spherical_j0 (#86049)

khushi-411 committed 2 years ago

d6b03085

[easy] Add spaces to vmap over as_strided error message (#86150)

zou3519 committed 2 years ago

8bce2f3d

delete special fake tensor new handling (#86144)

eellison committed 2 years ago

e1859c07

[5/N] [Dispatchable Collectives] Update send with CPU / CUDA implementations (#83859)

H-Huang committed 2 years ago

3f2e7d5c

Move ITT testing to its own test case (#86174)

jingxu10 committed 2 years ago

a75edfa9

Forward fix land race (unexpected successes) (#86186)

Chillee committed 2 years ago

b95e0fcc

Symbolic shapes mega merge PR (Oct 3) (#86160)

ezyang committed 2 years ago

79dd621f

Symintified factory functions (#86067)

Chillee committed 2 years ago

de752748

Batch of symintifications to allow more models to pass in inference (#86104)

Chillee committed 2 years ago

82d9592f

Stop modifying the global logger on `import functorch` (#86147)

zou3519 committed 2 years ago

a4ff07f1

Require bias to be contiguous for depthwise3x3_winograd backend (#85711)

soulitzer committed 2 years ago

fe190078

[maskedtensor] use masked_softmax for forward/backward instead of regular softmax (#85845)

george-qi committed 2 years ago

bc1d8840

.github: Improve sanity check for generated files (#86143)

seemethere committed 2 years ago

0db9419e

[GHF] Make EasyCLA unskippable (#86161)

malfet committed 2 years ago

5ca0f9e1

Unconditionally register Python decomps to Meta key in Python Dispatcher (#85750)

ezyang committed 2 years ago

f3d7ab54

Revert "Disable XLA test (#86123)" (#86151)

huydhn committed 2 years ago

06ddb1c0

Switch to checking EasyCLA on merge (#86127)

zpao committed 2 years ago

cda815dc

ANTIALIAS updated to Resampling.LANCZOS in torch/utils/tensorboard/summary.py (#85679)

GithubUserAccountAmazing committed 2 years ago

dfde7cf3

Newer Older