strisunshinewentingwang/pytorch

Pull Requests Commits

[sparsity] Add ability to keep sparsity parameters in modules (#66777)

z-a-f committed 2 years ago

e948856c

[Operator Versioning][Edge] Implement register function for upgrader (#67730)

cccclai committed 2 years ago

13faaff5

[AO] Clear the contents of the torch/ao/__init__.py (#69415)

z-a-f committed 2 years ago

4f5806de

[AutoAccept][Codemod][FBSourceClangFormatLinter] Daily `arc lint --take CLANGFORMAT`

CodemodService FBSourceClangFormatLinterBot committed 2 years ago

015e481a

Fixes mem_get_info when querying on a device other than the current device (#69640)

Mike Ruberry committed 2 years ago

dc87cf5f

[Vulkan] Thread-safe Vulkan backend for OSS (#69576)

beback4u committed 2 years ago

24d885f5

Reduce binary size of TensorCompare.cu (#68835)

ngimel committed 2 years ago

ecf9c82f

[Vulkan] Implement clone operator (#69551)

beback4u committed 2 years ago

3e560239

Run test_embedding_bag_with_no_grad_tensors only for TensorPipe (#69626)

pritamdamania committed 2 years ago

eb2a8034

Make make_dual redispatch (#68630)

soulitzer committed 2 years ago

b61c532f

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

Newer Older