strisunshinewentingwang/pytorch

Pull Requests Commits

Print stdout and stderr to console on parallel runs (#60869)

1ntEgr8 committed 3 years ago

c8fb7858

docs: add `permute` in torch docs (#60821)

ydcjeff committed 3 years ago

a8057e7e

[vulkan] Implement tanh activation function (#60695)

hanhsienhuang committed 3 years ago

d7c58e5a

[quant] Input-Weight Equalization - tests (#60378)

angelayi committed 3 years ago

da70dd19

[quant] Input-Weight Equalization - support for connected F.linear layer (#60272)

angelayi committed 3 years ago

dfb9c0ba

[quant] Input-Weight Equalization - support for connected linear layers (#60034)

angelayi committed 3 years ago

ddf2ce03

[quant] Input-Weight Equalization - support for F.linear layers (#59964)

angelayi committed 3 years ago

79173189

support non-contiguous tensor in bilinear (#38409)

joerg-de committed 3 years ago

387289d4

Make requires grad check run only when grad mode is enabled (#60740)

albanD committed 3 years ago

f118d20b

Add an optional Device parameter to pin_memory/is_pinned that does nothing (#60201)

ezyang committed 3 years ago

3ad3f20b

Remove some unnecessary functions from CUDAHooks (#59655)

ezyang committed 3 years ago

85af24f5

Port silu_backward to structured (#58661)

Freey0 committed 3 years ago

b52849b5

Make some comparisons explicit (#60505)

r-barnes committed 3 years ago

66f01db3

Enhance ProcessGroupWrapper with additional checks + refactor (#60237)

rohan-varma committed 3 years ago

f5341bd5

[torch/distributed] remove outdated FutureWarning in distributed/elastic/util/store.py (#60807)

Kiuk Chung committed 3 years ago

aaea81e3

Paren-matching kernel launch check without external deps (#60778)

r-barnes committed 3 years ago

94cdbbf4

Python error unit tests on delegation of backend_with_compiler_demo (#60689)

Amyh11325 committed 3 years ago

88b0518a

ENH Adds byte support for nll_loss (CUDA) (#60650)

thomasjpfan committed 3 years ago

e63db3ae

Add -I<directory> option to tools/linter/clang_tidy.py (#60745)

1ntEgr8 committed 3 years ago

7f6b2bc2

Don't reference reflection_pad3d in functional.py (#60837)

ngimel committed 3 years ago

5b118a7f

To add Nesterov Adam algorithm for multi-tensor optimizers API (#59165)

iramazanli committed 3 years ago

f0e972a4

[TensorExpr] Add a mechanism to register custom TS->NNC lowerings in TensorExprKernel. (#60804)

Mikhail Zolotukhin committed 3 years ago

3bfe1508

To add Rectified Adam algorithm for multi-tensor optimizers API (#59161)

iramazanli committed 3 years ago

5563f4bd

Support default values on NamedTuple fields (#54682)

Ansley Ussery committed 3 years ago

0fbc471d

fix cuda mem leak check not properly run on master_builds (#60742)

Rong Rong (AI Infra) committed 3 years ago

6b53792f

[Static Runtime] Remove output type constraints (#60669)

Hao Lu committed 3 years ago

e3abccec

Fix missing spaces in error of constant_pad_nd (#60729)

take-cheeze committed 3 years ago

dae25c20

Modernize for-loops in aten (#59598)

r-barnes committed 3 years ago

9a08e87d

supports non-leaf inputs for autograd.backward() function (#60521)

RockingJavaBean committed 3 years ago

7e3a694b

Remove un-used parameter in _trilinear backward (#60673)

albanD committed 3 years ago

056a8e0d

Newer Older