strisunshinewentingwang/pytorch

Pull Requests Commits

[mlf][efficiency] modify equalization scale operator to return single output (#46449)

hwangjeff committed 3 years ago

ecf63351

[TensorExpr] Correctly handle negative dimensions in aten::cat when lowering to tensor expressions. (#46446)

Mikhail Zolotukhin committed 3 years ago

4359c5e0

Remove variable_excluded_from_dispatch() check for factory functions. (#46371)

Ailing Zhang committed 3 years ago

ec5f81f9

fix type check for torch.quantization._numeric_suite (#46330)

Rong Rong committed 3 years ago

d1745c36

Add named tuple's error message and workaround for RET failure (#46347)

jwpark1985 committed 3 years ago

92921c82

Update VariableTypeManual.cpp to not use catchAllKernel. (#46353)

Ailing Zhang committed 3 years ago

d278e83e

Fixes bug in sspaddmm (#45113) (#45963)

vfdev-5 committed 3 years ago

dc7cd974

More Timer refinement (#46023)

Taylor Robie committed 3 years ago

dda95e69

Add Sigmoid operator from Caffe2 (#46286)

b-koopman committed 3 years ago

757173a4

[caffe2] Bypass memonger for in-place ops (#46378)

Hao Lu committed 3 years ago

16c52d91

Update default ouput extension in optimize_for_mobile.cc (#45598)

dreiss committed 3 years ago

faf03bd2

vmap: added fallback for in-place operators (#46191)

zou3519 committed 3 years ago

f96cb9de

Rename createLevelsBitset -> createVmapLevelsBitset; move it (#46190)

zou3519 committed 3 years ago

401c85b4

Add quick fix for view/inplace issue with DDP (#46406)

albanD committed 3 years ago

849bc77e

Remove codegen for old RegistrationDeclarations.h (#46370)

Ailing Zhang committed 3 years ago

ba92920a

Add operator benchmark for 4bit/8bit embedding lookups

bugra committed 3 years ago

03c7d5be

Fixing pow for special case between cuda tensors and cpu tensors and reframed test cases a tiny bit (#46320)

janeyx99 committed 3 years ago

c99378af

Migrate CPU_tensor_apply to TensorIterator (#44242)

muthuArivoli committed 3 years ago

7d6d5f4b

use @mode/ndk_libcxx instead of mode/gnustl

linbinyu committed 3 years ago

8f61fa65

Automated submodule update: FBGEMM (#46395)

facebook-github-bot committed 3 years ago

e69f2f82

Added support for complex torch.pinverse (#45819)

IvanYashchuk committed 3 years ago

c1141b6f

BFloat16 support for torch.sign (#45244)

zasdfgbnm committed 3 years ago

5ce46fbb

Add fence. (#45148)

Ashkan Aliabadi committed 3 years ago

8c26111a

Add image sampler. (#45037)

Ashkan Aliabadi committed 3 years ago

e3eef0cd

Redo Vulkan command and descriptor pools. (#44496)

Ashkan Aliabadi committed 3 years ago

50f83324

Fix error message for scatter reduction (#46397)

albanD committed 3 years ago

1e654a4b

Updated derivatives for complex mm, mv, ger, bmm, triangular_solve (#45737)

IvanYashchuk committed 3 years ago

528158af

Allow Undefined to get kernel from Math/DefaultBackend. (#46352)

Ailing Zhang committed 3 years ago

7f458e16

[JIT] Revert Freezing shared type PR (#46285)

Elias Ellison committed 3 years ago

908c2357

Add windows JNI support (#44257)

Elijah Rippeth committed 3 years ago

b5479737

Newer Older