strisunshinewentingwang/pytorch

Pull Requests Commits

Port nextafter to structured (#57625)

Freey0 committed 3 years ago

3a1dc60d

Port gcd to structured (#57624)

Freey0 committed 3 years ago

7e51ac5e

Fixing quantize_per_tensor on cuda (#57703)

HDCharles committed 3 years ago

5044d9dc

[Gradient Compression] Divide by world size before all_reduce to avoid overflow (#57410)

Weiyi Zheng committed 3 years ago

c07babbc

Copy edit of TorchScript Language Reference (#57694)

holly1238 committed 3 years ago

626ae7f0

Be more lenient with network exceptions in trigger_azure_pipeline.py (#57714)

skyline75489 committed 3 years ago

b5b158a6

[reland] Remove unused code in windows_build_definition.py (#57107)

skyline75489 committed 3 years ago

161ea537

add `torch.testing` to docs (#57247)

pmeier committed 3 years ago

0dd0151c

Fix test reporting regression (#57795)

malfet committed 3 years ago

27f672a0

make quantizeable MHA work with torch.jit.script (#57774)

vkuzo committed 3 years ago

2901d2e6

Revert D28248766: Update internal code for torch.linalg.solve

samestep committed 3 years ago

023ecc40

New: sparse complex: add linear algebra, addmm (#57129)

aocsa committed 3 years ago

6f2c0ccc

New: Initial support for sparse complex tensors constructors for CPU/CUDA (#57125)

aocsa committed 3 years ago

a911c4fc

[FX] Adds PyTree support to FX through `concrete_args` (#55888)

Chillee committed 3 years ago

8d363d37

Migrate from shared_ptr to intrusive_ptr for Future (#57636)

lw committed 3 years ago

45012da2

Pass reference to parent future in callbacks (#57635)

lw committed 3 years ago

36e47af5

Make wrapPropagateTLSState more generic (#57634)

lw committed 3 years ago

9aa1461a

Update internal code for torch.linalg.solve (#56613)

IvanYashchuk committed 3 years ago

5f292507

Update internal code for at::_lu_with_info (#56612)

IvanYashchuk committed 3 years ago

adaf80bc

OpInfo: expand and expand_as (#57606)

kshitij12345 committed 3 years ago

9e6b7e6e

[testing] clean-up test_unary_ufuncs.py (#57615)

kshitij12345 committed 3 years ago

1f7309df

OpInfo: float_power (#57648)

krshrimali committed 3 years ago

4cb3c60c

[testing] atan2: Enable cases where self broadcasts (#57608)

kshitij12345 committed 3 years ago

6eec730a

fft: Increase tolerance for nd-fft tests (#57576)

peterbell10 committed 3 years ago

159a2404

[testing] change unaryufunc default dtypes (#57616)

kshitij12345 committed 3 years ago

ee79413b

Add call_kwargs(args, kwargs) method to torch::deploy api (#57748)

wconstab committed 3 years ago

319b08be

[iOS GPU][Perf][1/n] Use aten::contiguous instead of permuting weights manually (#57664)

xta0 committed 3 years ago

f2fdb61e

[Pytorch Edge] Enable eager symbolication in benchmarking binary (#57705)

kimishpatel committed 3 years ago

ca8090f8

[torch/elastic] Rename etcd-/c10d-experimental to etcd-v2 and c10d (#57764)

cbalioglu committed 3 years ago

e5e095cb

Automated submodule update: FBGEMM (#57485)

facebook-github-bot committed 3 years ago

cb95c9db

Newer Older