strisunshinewentingwang/pytorch

Pull Requests Commits

Adding hardswish, opinfo tests to custom rules (#69399)

Gamrix committed 2 years ago

adb619a1

[Operator Versioning][Edge] Have operator version number available at the loading stage (#67729)

cccclai committed 2 years ago

a0efa48c

Forward fix for failing master (#69625)

anjali411 committed 2 years ago

2808563e

Add efficient zero tensors (#64837)

anjali411 committed 2 years ago

3e616444

Add nvidia-smi memory and utilization as native Python API (#69104)

vincentpierre committed 2 years ago

30bb4e00

Improve efficiency of shape hash by not using tostring (#69496)

wconstab committed 2 years ago

ee60b5dd

OpInfo for `nn.functional.dropout2d`, revise sample inputs for `dropout` (#67891)

krshrimali committed 2 years ago

2cb385dd

add `OpInfo` for `torch.diagflat` (#65680)

pmeier committed 2 years ago

f54745a6

add `OpInfo` for `torch.nn.functional.kl_div` (#65469)

pmeier committed 2 years ago

7e49f463

add python dispatch test back to CI and fix typo in test (#69565)

albanD committed 2 years ago

8b20dde9

[Static Runtime] Avoid evaluating expressions of `Node*` for interpreter fallback op (#69489)

d1jang committed 2 years ago

afaa184b

Updating quantization documentation (#68907)

HDCharles committed 2 years ago

fc261453

[DataLoader] Implementing communication processes for Map-style DataPipes (#68549)

NivekT committed 2 years ago

39fb855d

[quant][embdding qat] Re-land Add FX support for QAT EmbeddingBag (#69334)

b-koopman committed 2 years ago

f3983f9c

[quant][embedding qat] Re-Land Support Embedding QAT via FX API (#69333)

b-koopman committed 2 years ago

93aa3603

histc: Avoid dispatch in parallel region (#68520)

peterbell10 committed 2 years ago

fc8404b5

Fix TSAN issue in TCPStore (#69590)

pritamdamania committed 2 years ago

2a38e1a7

Release GIL during RPC shutdown. (#69586)

pritamdamania committed 2 years ago

0ce49000

OpInfo tests for `(svd|pca)_lowrank` (#69107)

nikitaved committed 2 years ago

c2362478

Fix sign op converter (#69580)

Shirong Wu committed 2 years ago

e06af791

Remove finput and fgrad_input from slow3d transpose signatures (#68899)

jbschlosser committed 2 years ago

6b950eea

[quant][graphmode] initial support for fusion pattern in backend_config_dict (#69335)

jerryzh168 committed 2 years ago

05946051

use irange for loops (#69533)

r-barnes committed 2 years ago

2d38d37f

[LT] Sync with the lazy_tensor_staging branch (#69527)

desertfire committed 2 years ago

8a975c01

[Reland][Autograd/Checkpoint] Checkpoint implementation without reentrant autograd (#69508)

rohan-varma committed 2 years ago

049debd9

Allow build_android.sh to forward Vulkan args (#69332)

SS-JIA committed 2 years ago

3456c2cb

[vulkan] Disable shader optimization to avoid Validation Errors (#69331)

SS-JIA committed 2 years ago

fa39754e

[vulkan] Add image format qualifier to glsl files (#69330)

SS-JIA committed 2 years ago

bede33e3

[quant][graphmode] Refactor fusion to use the new Pattern format (#68770)

jerryzh168 committed 2 years ago

e5a1ee0e

use irange for loops 6 (#66742)

r-barnes committed 2 years ago

1433160a

Newer Older