strisunshinewentingwang/pytorch

Pull Requests Commits

Improve documentation and warning message for creation of a tensor with from_numpy() (#49516)

leonvol committed 3 years ago

4a6c178f

Revert D25757721: [pytorch][PR] Run mypy on more test files

Mike Ruberry committed 3 years ago

9529ae37

[docs] add docstring in torch.cuda.get_device_properties (#49792)

Jeff Yang committed 3 years ago

d1a56fcd

[JIT] Add `__prepare_scriptable__` duck typing to allow replacing nn.modules with scriptable preparations (#45645) (#49242)

Meghan Lele committed 3 years ago

abe1fa49

[pytorch][codegen] migrate gen_variable_type to new data model (#49735)

ljk53 committed 3 years ago

e71a13e8

[PyTorch] Avoid heap allocations in inferUnsqueezeGeometry (#49497)

swolchok committed 3 years ago

a272a7ee

Enable distribution validation if __debug__ (#48743)

fritzo committed 3 years ago

093aca08

Revert D25757691: [pytorch][PR] Run mypy over test/test_utils.py

Rong Rong (AI Infra) committed 3 years ago

e3c56ddd

Update MultiHeadAttention docstring (#49950)

vfdev-5 committed 3 years ago

e442ac1e

Drop unused imports from caffe2/python (#49980)

r-barnes committed 3 years ago

9945fd72

[caffe2][a10] Move down pragma pop to properly suppress warning 4522 (#49233)

jdonald committed 3 years ago

eee849be

Fix store based barrier to only use 'add'. (#49930)

pritamdamania committed 3 years ago

16e5af41

support building with conda installed libraries (#50080)

Rong Rong (AI Infra) committed 3 years ago

12ee7b61

[RPC] Relax some profiling tests (#49983)

rohan-varma committed 3 years ago

e868825e

[distributed] Provide parameter to pass GPU ID in barrier function (#49069)

jaglinux committed 3 years ago

c115957d

Add an option to disable aten::cat in TE (re-revert) (#50101)

Krovatkin committed 3 years ago

3cd2f1f3

[JIT] Remove buffer metadata serialization forward-compat gate (#49990)

Meghan Lele committed 3 years ago

bbae6774

eager quant: fix error with removing forward hooks (#49813)

vkuzo committed 3 years ago

04e86be1

quant: throw a nice error message for allclose with quantized inputs (#49802)

vkuzo committed 3 years ago

113b7623

quant: nice error message on convtranspose with per-channel weight (#49899)

vkuzo committed 3 years ago

44c17b28

quant: ensure observers do not crash for empty Tensors (#49800)

vkuzo committed 3 years ago

72306378

Run mypy over test/test_utils.py (#49654)

rgommers committed 3 years ago

c86cfcd8

Run mypy on more test files (#49658)

rgommers committed 3 years ago

b7bfc723

fixes indices computation for trilinear interpolate backwards (#50084)

ngimel committed 3 years ago

e35b822d

Patch death tests/fork use after D25292667 (part 3)

andriigrynenko committed 3 years ago

52933b99

Revert D25763758: [pytorch][PR] introduce a flag to disable aten::cat in TE

heitorschueroff committed 3 years ago

ace78ddb

Fixing error in Readme.md. (#50033)

yqtianust committed 3 years ago

38457703

Fix grammar typo in readme.md (#50000)

yqtianust committed 3 years ago

8c66aec4

Fix return value of _vmap_internals._get_name (#49951)

alexhenrie committed 3 years ago

e4d596c5

unit test for fc parallelization aot (#50056)

Hector Yuen committed 3 years ago

6e6231f9

Newer Older