strisunshinewentingwang/pytorch

Pull Requests Commits

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

CMake: Support dynamic codegen outputs (#68246)

peterbell10 committed 2 years ago

9a7732e8

Rationalize API exports in torch_python (#68095)

peterbell10 committed 2 years ago

cd9da326

Output UnionType str rep with () instead of [] (#69502)

tugsbayasgalan committed 2 years ago

829b49b8

Sparse CSR CUDA: Add block torch.addmv when mat is sparse (#68708)

IvanYashchuk committed 2 years ago

a8232ee1

skip ORT tensor in TensorIterator because it doesn't have storage (#68705)

souptc committed 2 years ago

6df7b751

[SR] Simplify memory re-use algorithm (#68302)

Mike Iovine committed 2 years ago

008469c5

Making cuda 11.5 workflows periodic (#69323)

atalman committed 2 years ago

c3096379

Add conda-forge dependency for cuda-11.5 (#69541)

malfet committed 2 years ago

baac51ff

Add Union type to TorchScript Language Ref (#69514)

gmagogsfm committed 2 years ago

358e9081

[JIT] optimize_for_inference on methods other than forward (#69367)

davidberard98 committed 2 years ago

c21169ea

[JIT] run frozen optimizations on methods other than forward (#68668)

davidberard98 committed 2 years ago

60ca6776

Sparse CSR: Implement unary ufuncs (with 0->0 correspondence) (#69292)

krshrimali committed 2 years ago

63470f94

Docs: Fix broken code syntax in autograd.rst (#69362)

rodrigobdz committed 2 years ago

1a202b0c

trt engine inspector demo (#66683)

Yinghai Lu committed 2 years ago

10229e15

[JIT] check stack size after calling operator (#68788)

davidberard98 committed 2 years ago

aa9fbb9a

Older