strisunshinewentingwang/pytorch

Pull Requests Commits

Make make_dual also return namedtuple when level less than zero (#68628)

soulitzer committed 2 years ago

7956a405

[SR] Scope exit guard for memory planner deallocation (#68795)

Mike Iovine committed 2 years ago

1c43b160

Fix typos in ATen README (#69170)

Nikita Karetnikov (ニキータカレートニコフ) committed 2 years ago

3b27304d

Port smooth_l1_loss to structured kernels (#67404)

Jiewen Tan committed 2 years ago

b10381f4

Getting NS to work with Ferraris (#68908)

HDCharles committed 2 years ago

497ec9d9

[PyTorch Tests] Split out skip logic, make changes for plugins (#67256)

bryanmr committed 2 years ago

51b6981c

Remove remaining THC code (#69039)

peterbell10 committed 2 years ago

e279963e

[fix] cross_entropy : fix weight with ignore_index and label_smoothing (#69511)

kshitij12345 committed 2 years ago

7407e3d6

[BE] Enable C++ stacktraces for MultiProcessTestCase (#69175)

rohan-varma committed 2 years ago

d44d59aa

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

Newer Older