strisunshinewentingwang/pytorch

Pull Requests Commits

[Gradient Compression] Do not skip the comm hook tests for Gloo/MPI backends (#58444)

Yi Wang committed 3 years ago

df44f015

Conservatively move all suitable prim ops from full-jit to mobile, and make them selective. (#58353)

iseeyuan committed 3 years ago

c3861649

[Pytorch] Build lite interpreter as default for iOS

cccclai committed 3 years ago

b5a834a7

Wrap torch::deploy API functions in safe rethrow macros (#58412)

wconstab committed 3 years ago

8a3fb268

[FX] Fix retracing wrapped functions (#58061)

James Reed committed 3 years ago

7b73fdf5

Make new_ones an operator (#58405)

Chillee committed 3 years ago

5fa4541c

Change link order for BUILD_SPLIT_CUDA option (#58437)

malfet committed 3 years ago

0547a3be

Add shape documentation for CosineEmbeddingLoss (#58403)

jbschlosser committed 3 years ago

af463d22

add kernel launch checks after each kernel launch to silence the check (#58432)

ngimel committed 3 years ago

e24dee00

[package] fix persistent_load error (#58439)

Lilyjjo committed 3 years ago

7dd08504

Clang format distributed_c10d.py (#58435)

Yi Wang committed 3 years ago

314a5781

[ATen] Use MaybeOwned<T> in at::argmin/argmax (#58338)

Hao Lu committed 3 years ago

b6d3929b

Remove timeouts for C2 tests

Taylor Robie committed 3 years ago

6989eb60

.github: Add intial Windows CPU GHA workflow (#58199)

seemethere committed 3 years ago

4310decf

fx quant: fix crash on output dicts and lists (#58416)

vkuzo committed 3 years ago

c156a4ff

fx quant: remove test debug logs (#58415)

vkuzo committed 3 years ago

a1cacf3b

[ONNX] Fix split export in opset13 (#56277) (#57605)

BowenBao committed 3 years ago

3d12ab45

[Pytorch] Build lite interpreter as default for Android

cccclai committed 3 years ago

0c3db1cb

[torch] Format repeat_interleave op files (#58313)

Serhat Yilmaz committed 3 years ago

d645088f

Merge CreationMeta MULTI_OUTPUT_SAFE with MULTI_OUTPUT_NODE (#58285)

soulitzer committed 3 years ago

06c1094e

Remove unused python2 shebang (#58409)

samestep committed 3 years ago

3507ca32

Use torch.allclose to check tensor equality (#58429)

malfet committed 3 years ago

98cc0aa6

Enable NNAPI in internal build (#58324)

dreiss committed 3 years ago

50f9a181

Don't bind Android NNAPI on Apple platforms (#58323)

dreiss committed 3 years ago

532632ca

[Pytorch] Remove run_on_bundled_input (#58344)

JacobSzwejbka committed 3 years ago

1891e4bf

Improve error message when Proxy object is iterated (#58302)

Ansley Ussery committed 3 years ago

443ce1e8

Chown workspace in calculate-docker-image (#58398)

samestep committed 3 years ago

a4ce85ad

Improve `CONTRIBUTING.md` (#58396)

Ansley Ussery committed 3 years ago

e8981e7c

Reland OpInfo support for forward AD (#58304)

albanD committed 3 years ago

9afe9fba

generate inplace/out kernels for xla (#57510)

bdhirsh committed 3 years ago

1a9efbbc

Newer Older