strisunshinewentingwang/pytorch

Pull Requests Commits

[Gradient Compression] Provide 2 default C++ comm hooks (#46701)

Yi Wang committed 3 years ago

e299393f

[Gradient Compression] Add CppCommHook subclass for supporting the C++ API of communication hook. (#46566)

Yi Wang committed 3 years ago

e077a2a2

[quant][graphmode][fx] custom_module support static/dynamic/weight_only quant (#46786)

jerryzh168 committed 3 years ago

998b9b9e

[quant][graphmode][fx][fix] scalar as first input for add/mul (#46751)

jerryzh168 committed 3 years ago

5a8198eb

[OpBench] fix jit tracing with quantized op/tensor by enabling `_compare_tensors_internal` to compare quantized tensors (#46772)

Yang Wang committed 3 years ago

810c68fb

Added foreach_zero_ API (#46215)

izdeby committed 3 years ago

8e37dcb1

[FX] Fix handling of `inf` and `nan` literals (#46894)

James Reed committed 3 years ago

67c1dc65

Fix internal assert for torch.heaviside with cuda tensor and cpu scalar tensor (#46831)

Kiyosora committed 3 years ago

53839ac9

[caffe2] Disable running full grad check in tests by default

Dmytro Dzhulgakov committed 3 years ago

115bbf99

quant: fix bug with copy.deepcopy of FX prepared quantization models (#46895)

vkuzo committed 3 years ago

8066e89f

Add CUDA 11.1 CI (#46616)

zasdfgbnm committed 3 years ago

1479ed91

Install sccache from source (#46672)

malfet committed 3 years ago

c20c840c

Adding link to gcov depending on GCC_VERSION (#46928)

janeyx99 committed 3 years ago

64d4b24a

Conditional requirement for py3.6 only (#46932)

ccoulombe committed 3 years ago

dc53eefd

[iOS] Bump up the cocoapods version (#46935)

husthyc committed 3 years ago

79a1d2bd

add type annotations to comm.py (#46736)

guilhermeleobas committed 3 years ago

717e6d80

remove event not ready assertion from TestCuda.test_copy_non_blocking (#46857)

jeffdaily committed 3 years ago

151f31ba

Fix typo in setup.py (#46921)

malfet committed 3 years ago

8c39f198

[numpy] `torch.log{2,10}` : promote integer inputs to float (#46810)

kshitij12345 committed 3 years ago

21e60643

Add GradMode::enabled check to max_pool1d (#46767)

heitorschueroff committed 3 years ago

bbe5bfaa

Increase no-output-timeout for OSX builds (#46891)

malfet committed 3 years ago

daf2a6a2

Update dper3 to use torch.nan_to_num and nan_to_num_ (#46873)

ShijunK committed 3 years ago

d5cd781c

add sparse_nn_partition (#46390)

scottxu0730 committed 3 years ago

86409050

Replace flatten tensors with flatten loops. (#46737)

navahgar committed 3 years ago

4b6e3071

[quant][graphmode][fx] Support sigmoid/hardsigmoid/tanh in qat (#46738) (#46871)

jerryzh168 committed 3 years ago

6b50ccc4

Add single element tuple output from to_backend/to_glow (#5029)

jackm321 committed 3 years ago

60eded6c

Add a warning message that torch.sign would not support complex numbers (#43280)

xuhdev committed 3 years ago

bcbb6bac

add fburl link to error message (#46795)

linbinyu committed 3 years ago

37da6d26

Fix TripletMarginWithDistanceLoss example code (#46853)

soulitzer committed 3 years ago

9858b012

[c10] fix weak_intrusive_ptr lock() (#46007)

wanchaol committed 3 years ago

4a35280e

Newer Older