strisunshinewentingwang/pytorch

Pull Requests Commits

reenabling MPI test (#48725)

Rong Rong committed 3 years ago

ef50c94e

Replace constexpr with CONSTEXPR_EXCEPT_WIN_CUDA (#48717)

datumbox committed 3 years ago

0484b048

Add support for CorrCholeskyTransform (#48041)

neerajprad committed 3 years ago

5489a98c

Add broadcast_shapes() function and use it in MultivariateNormal (#43935)

fritzo committed 3 years ago

313e77fc

Revert D24874754: [pytorch][PR] Add test for empty tensors for batch matmuls

Mike Ruberry committed 3 years ago

c7746adb

Optimize torch zeros (#45636)

Edvard Ghazaryan committed 3 years ago

79b9c034

Fix unintended error when worker force kill happens #43455 (#43462)

ssnl committed 3 years ago

1112773c

Replace kernel resource strings with real .cu source files (#48283)

tlemo committed 3 years ago

85c1e8ac

Add test for empty tensors for batch matmuls (#47700)

zasdfgbnm committed 3 years ago

5f105e2a

[qunat][graphmode][fx] Standalone module takes float as input and output (#48671)

jerryzh168 committed 3 years ago

ea573ea9

Disable autocast cache for tensor views as fix for #48049 (#48696)

pbialecki committed 3 years ago

22c3ae8b

Refactored OpInfo testing to support custom SampleInputs, added addmm to op_db to test (#48627)

Lilyjjo committed 3 years ago

0e4f9a78

Support for OpInfo-based testing for operators in JIT (#47696)

Lilyjjo committed 3 years ago

90faf431

Refactored assertAutodiff test to have better error message (#48567)

Lilyjjo committed 3 years ago

9c35a680

Refactor existing JIT testing utils to enable new OpInfo test suite to reuse existing logic (#47695)

Lilyjjo committed 3 years ago

c465602d

[NNC] Add cpu fusion gflag (#48682)

Elias Ellison committed 3 years ago

11954039

only enable cudnn persistent RNN when batchsize % 8 == 0 (#48070)

xwang233 committed 3 years ago

0d39bd47

Use fastAtomicAdd in GPU upsampling trilinear (#48675)

xwang233 committed 3 years ago

a1daf1e6

Hipify revamp [REDUX] (#48715)

jithunnair-amd committed 3 years ago

5f623087

torch: Stop using _nt_quote_args from distutils (#48618)

seemethere committed 3 years ago

780f2b9a

Vulkan linear memory allocator. (#48569)

Ashkan Aliabadi committed 3 years ago

95311add

[fix] repr(torch.device) (#48655)

kshitij12345 committed 3 years ago

90a3049a

Add reparameterization support to `OneHotCategorical` (#46610)

lqf96 committed 3 years ago

b006c7a1

[vulkan] Distribute weight prepacking along y dimension for conv2d (#48266)

SS-JIA committed 3 years ago

de46369a

add type annotations to common_nn.py (#48190)

guilhermeleobas committed 3 years ago

a4e13fcf

Remove "-b" option from `pip install` command (#48742)

malfet committed 3 years ago

a49e2c5c

[JIT] Fix clang-tidy warnings in jit/runtime (#47992)

Meghan Lele committed 3 years ago

fc1153a8

[JIT] Fix clang-tidy warnings in jit/serialization (#47991)

Meghan Lele committed 3 years ago

a25d52f4

[JIT] Fix clang-tidy warnings in jit/testing (#47986)

Meghan Lele committed 3 years ago

34b2304e

[JIT] Fix clang-tidy warnings in jit/python (#47985)

Meghan Lele committed 3 years ago

18eccfbe

Newer Older