strisunshinewentingwang/pytorch

Pull Requests Commits

Disable NEON acceleration on older compilers (#47099)

malfet committed 3 years ago

c86af4aa

[quant][graphmode][fx][fusion] Add test for fuse_fx (#47085)

jerryzh168 committed 3 years ago

085193c2

Add C++ coverage for Ubuntu cpu tests (#46656)

janeyx99 committed 3 years ago

1dd220bd

Fix mul cuda for bool (#47031)

malfet committed 3 years ago

edac4060

use bitfield to shrink TensorImpl (#45263)

Basil Hosmer committed 3 years ago

69fe10c1

faster TensorOptions merging (#45046)

Basil Hosmer committed 3 years ago

99fed7bd

track Half/ComplexHalf default dtype (#45043)

Basil Hosmer committed 3 years ago

c7fc8cab

pass TypeMeta by value (#45026)

Basil Hosmer committed 3 years ago

f05b66b7

Fix max_pool2d with ceil_mode bug (#46558)

heitorschueroff committed 3 years ago

26438008

Automated submodule update: FBGEMM (#47071)

facebook-github-bot committed 3 years ago

7df0224c

add warning if DataLoader is going to create excessive number of thread (#46867)

lixinyu committed 3 years ago

67b7e751

Add last_n_window_collector

bugra committed 3 years ago

eec201c1

add add_node function for partition to fix partition mem size calculation (#47083)

scottxu0730 committed 3 years ago

6c34aa72

[JIT] Add selective backend lowering API (#43613)

Meghan Lele committed 3 years ago

f9d32c4f

Split comm hooks into python-dependent hooks and others (#47019)

gmagogsfm committed 3 years ago

0dbd7293

[pytorch] add script to run all codegen (#46243)

ljk53 committed 3 years ago

d95e1afa

Fix links in tools/build_variables.bzl (#47066)

samestep committed 3 years ago

707d2714

[quant][graphmode][fx] Remove logging for standalone module api calls (#47032)

jerryzh168 committed 3 years ago

366888a5

[pytorch/ops] Concat fast path w/ zero tensor (#46805)

xw285cornell committed 3 years ago

e3b55a8a

Fix lint (#47095)

James Reed committed 3 years ago

2e2dc587

Enable python code coverage on windows (#44548)

guyang3532 committed 3 years ago

f5477e37

Fix median bug on discontigous tensors (#46917)

heitorschueroff committed 3 years ago

ddeacf15

[WIP] Move torch.fx into its own target (#46658)

James Reed committed 3 years ago

9bc8f071

[Transposed Conv]add ConvTranspose3d with FBGEMM as backend (#46608)

YazhiGao committed 3 years ago

71901554

Pin destination memory for cuda_tensor.to("cpu", non_blocking=True) (#46878)

mcarilli committed 3 years ago

3c643d11

fix calculation of number of elements to not overflow (#46997)

ngimel committed 3 years ago

e17b8dea

Replace -f with -x for pytest tests. (#46967)

pritamdamania committed 3 years ago

78de12f5

[ONNX] bump CI ort to 1.5.2 rel for stability (#46595)

BowenBao committed 3 years ago

a4caa3f5

Delete TypeDefault.h and TypeDerived.h codegen entirely. (#47002)

ezyang committed 3 years ago

843cab3f

Delete TypeDefault call code generation logic in VariableType (#47000)

ezyang committed 3 years ago

c689b4d4

Newer Older