strisunshinewentingwang/pytorch

Pull Requests Commits

Remove .impl_UNBOXED() and functionalities associated with it (#49220)

smessmer committed 3 years ago

4a14020c

Remove codegen logic to support non-c10-full ops (#49164)

smessmer committed 3 years ago

e4c41b69

Add android.permission.INTERNET permission to Android test_app. (#49996)

Ashkan Aliabadi committed 3 years ago

fcb69d2e

Remove redundant code for unsupported Python versions (#49486)

hugovk committed 3 years ago

473e78c0

[vulkan] 2D prepacking for conv2d (#48816)

SS-JIA committed 3 years ago

09eb4683

[PyTorch Mobile] Generate Kernel dtype selection code in selected_mobile_ops.h during the build (#49279)

dhruvbird committed 3 years ago

9b519b4a

Remove incorrect links to zdevito/ATen (#50065)

samestep committed 3 years ago

ba691e1a

[JIT] disable masked fill (#50147)

Elias Ellison committed 3 years ago

6eee2a0a

Back out "Revert D25757721: [pytorch][PR] Run mypy on more test files" (#50142)

ezyang committed 3 years ago

3ce53988

Clean up type annotations in torch/nn/quantized/modules (#49941)

r-barnes committed 3 years ago

63808695

Implementation of torch::cuda::synchronize (#50072)

jbschlosser committed 3 years ago

7d9eb6c6

[Needs Review] Convert some files to Python3 (#49351)

r-barnes committed 3 years ago

e606e603

Clean up some type annotations in torch/testing/_internal (#50078)

r-barnes committed 3 years ago

efe0533a

Fix mypy type hint for AdaptiveAvgPool2,3d, AdaptiveMaxPool2,3d (#49963)

VinhLoiIT committed 3 years ago

74c055b2

Push anonymous namespace into codegen, not template (#49498)

ezyang committed 3 years ago

68a6e463

[PyTorch] IValue::toTensor can now return const Tensor& (#48868)

swolchok committed 3 years ago

480a7561

[PyTorch] Store Tensor explicitly in IValue (#48824)

swolchok committed 3 years ago

1b31e135

[PyTorch] Additional IValue tests (#49718)

swolchok committed 3 years ago

688992c7

Unused variables in neural net classes and functions (#50100)

alexhenrie committed 3 years ago

5f2ec629

Add support for converting sparse bool tensors to dense (#50019)

NathanHowell committed 3 years ago

c517e15d

Fix cl.exe detection in cpu/fused_kernel.cpp (#50085)

skyline75489 committed 3 years ago

2ac180a5

Set USE_RCCL cmake option (dependent on USE_NCCL) [REDUX] (#34683)

jithunnair-amd committed 3 years ago

45ec3582

drop a unneeded comma from cmakelist.txt (#50091)

cyyever committed 3 years ago

0ad6f066

Revert D25239967: [fx] Add matrix multiplication fusion pass

ngimel committed 3 years ago

ad7d208b

[PyTorch] Reapply D25546409: Use .sizes() isntead of .size() in cat_serial_kernel_impl (#49762)

swolchok committed 3 years ago

282552dd

Fix for possible RNG offset calculation bug in cuda vectorized dropout with VEC=2 (#50110)

mcarilli committed 3 years ago

57d489e4

[reland][quant][graphmode][fx] Standalone module support {input/output}_quantized_idxs (#49754) (#50058)

jerryzh168 committed 3 years ago

f6f0fde8

Fix mypy typing check for test_dataset (#50108)

ejguan committed 3 years ago

05358332

Remove cpu half and dead code from multinomial (#50063)

t-vi committed 3 years ago

def8aa54

[fx] Add matrix multiplication fusion pass (#50120)

Meghan Lele committed 3 years ago

9b7f3fa1

Newer Older