strisunshinewentingwang/pytorch

Pull Requests Commits

[iOS GPU][Perf][3/n] Cache the compuation pipeline state object (#57666)

xta0 committed 3 years ago

f3e014f3

[fx ir] Handle the case when output consumes get_attr directly (#57844)

khabinov committed 3 years ago

36a22967

Alias det, slogdet, matrix_power, inverse, pinverse (#57821)

lezcano committed 3 years ago

a93314de

Deprecate torch.lstsq (#57743)

lezcano committed 3 years ago

ba84c911

[nccl] log rank when communicator is aborted (#57974)

rohan-varma committed 3 years ago

5840c8cf

OpInfo: select (#57731)

kshitij12345 committed 3 years ago

a0d686c9

[c10d] Log when store based barrier succeeds (#57711)

rohan-varma committed 3 years ago

e90fcffb

Only use actually mismatched elements for reporting in `torch.testing` (#57923)

pmeier committed 3 years ago

71ca3e99

[kineto] Update Kineto submodule, cupti library paths (#57789)

Ilia Cherniavskii committed 3 years ago

c7145960

[nnc] Fix float->bool conversion on cpu (#57798)

bertmaher committed 3 years ago

f97650e7

Add tests for custom state_dict save/load methods in TorchScript (#57886)

gmagogsfm committed 3 years ago

b8ca1219

Add enabling default instructions flag for mobile (#57778)

tugsbayasgalan committed 3 years ago

fc9c4860

[RPC Framework] Move the annotation w/ bold effect out of the quotes (#57965)

Yi Wang committed 3 years ago

38500d5d

Support for accumulate nodes traversal and to access op names in the compare function (#57685)

Aravind Kalaiah committed 3 years ago

747312bf

Revert D28294662: [pytorch][PR] add cuda memory and distributed metadata

albanD committed 3 years ago

03616711

[iOS GPU][Perf][2/n] Prepack linear + Fuse relu/hardtanh (#57665)

xta0 committed 3 years ago

36172f34

Disable test for quicklint (#57968)

davidriazati@fb.com committed 3 years ago

f1d01b94

torch.where is now mentioning Bool rather than Byte when given wrong dtype mask (#57942)

venkateshtata committed 3 years ago

f0f69c5d

Fix json parse error for profiler call stack (#57099)

guyang3532 committed 3 years ago

ebb1b74f

add cuda memory and distributed metadata (#57252)

guotuofeng committed 3 years ago

98fcdb80

Fix typo in warning for spawn method (#57927)

kaushikb11 committed 3 years ago

ba07aaf2

[vulkan] Add sigmoid activation functions (#57867)

SS-JIA committed 3 years ago

19706d91

Fix creation_meta for multi view outputs in NoGradMode/InferenceMode. (#57842)

Ailing Zhang committed 3 years ago

481806be

[Vulkan] Fix seg fault during descriptor set allocation on some platforms (#57825)

SS-JIA committed 3 years ago

478f6397

Use tools/print_test_stats.py in GHA (#57647)

samestep committed 3 years ago

e1cbc43f

Fix hasattr support type (#57950)

nikithamalgi committed 3 years ago

bf053a12

Ensure torch.save() deterministic output (#57536)

FrancescoCasalegno committed 3 years ago

fea38242

[DDP] fix param to name mapping (#57771)

rohan-varma committed 3 years ago

fe3c63d9

Add Python-3.9 CI testing (#50992)

malfet committed 3 years ago

b587354e

Do not download slow test when on sandcastle (#57953)

Rong Rong (AI Infra) committed 3 years ago

29753339

Newer Older