strisunshinewentingwang/pytorch

Pull Requests Commits

[jit][subgraph_matcher] Enable regex matching for string attributes of node (#39454)

jerryzh168 committed 4 years ago

3669e457

[jit] update to serialization doc (#39025)

wanchaol committed 4 years ago

85621550

[online trainer] Add blob reorder (#39534)

ChunliF committed 4 years ago

83456923

disable autograd while preparing Tensor for printing (#39420)

albanD committed 4 years ago

e29d873e

observer bench: add CUDA (#39360)

vkuzo committed 4 years ago

e35199a6

Remove test_nccl from ROCM_BLACKLIST and enable only a couple of test_nccl tests (#39354)

jithunnair-amd committed 4 years ago

545a3e1e

reduce number of bailout nodes (#38281)

Krovatkin committed 4 years ago

97a2918a

[Docs] Update torch.(squeeze, split, set_printoptions, save) docs. (#39303)

KushajveerSingh committed 4 years ago

88fe05e1

Support torch.Tensor subclass (like Parameter) input. (#39487)

Yanan Cao committed 4 years ago

0031108b

fix input schema check for spatialbn

Hector Yuen committed 4 years ago

a6690bdb

Fix IDE hint channels_last & preserve_format (#39120)

Kiyosora committed 4 years ago

51504cb8

Un-inline Functions.h into Functions.cpp (#39446)

ezyang committed 4 years ago

77798a45

Update cafe2 hypothesis_test_util to support hypothesis-5 (#39498)

malfet committed 4 years ago

e2a178ca

Release GIL when deleting users and unforked owners (#39555)

mrshenli committed 4 years ago

baf6ed02

Fix possible deadlock in _wait_all_workers (#39535)

lw committed 4 years ago

9bfb91b5

Add @rpc.functions.async_execution for rpc.remote (#39486)

mrshenli committed 4 years ago

8a6914dd

Make @rpc.functions.async_execution processing generic (#39485)

mrshenli committed 4 years ago

11abb753

Explicitly decref in UnpickledPythonCall dtor (#38398)

Shen Li committed 4 years ago

fa4ed171

Refactor unittests for activation functions relu, elu, and sigmoid (#39190)

Paul Shao committed 4 years ago

876b9591

Bumps supported file format in anticipate of torch.div changes (#39529)

Mike Ruberry committed 4 years ago

7d56ef27

Added Operator_Schema's for missing FakeFP16 Operators (#39363)

venkatacrc committed 4 years ago

17aebe90

[quant][graphmode] Test for another type of ops in insert_observer for if (#39380)

jerryzh168 committed 4 years ago

f94a171e

Remove useless copy on zip file load (#36362)

davidriazati committed 4 years ago

da8191a9

misc updates to fake fp16 tests (#39405)

venkatacrc committed 4 years ago

ed12df64

Do not raise decorator (#39532)

malfet committed 4 years ago

2a513a6a

Reduce time spent per guard by comparing TensorType with Tensor (#39098)

Ailing Zhang committed 4 years ago

b861daf0

Add/fix typing annotations to some functions (#39075)

malfet committed 4 years ago

8811e4d0

deepcopy() of Objects should call __g/setstate__ (#39500)

James Reed committed 4 years ago

da2f8c9f

[ONNX] Fix type casting for reduce ops (#38829)

neginraoof committed 4 years ago

4e5af8d1

Upgrade lint. (#39483)

ezyang committed 4 years ago

da2004e1

Newer Older