strisunshinewentingwang/pytorch

Pull Requests Commits

Automated submodule update: FBGEMM (#57983)

facebook-github-bot committed 3 years ago

86b7ae18

[pyper] allow static runtime on and glow on simultaneously (#57972)

ajyu committed 3 years ago

eb1ffa91

Adding support for normalization of __is__ op (#57862)

Gamrix committed 3 years ago

698be312

Revert D28338485: make ddp logging api to be private

kimishpatel committed 3 years ago

ad4cd6ef

[local lint] Force color output on mypy (#58071)

driazati committed 3 years ago

a0230592

Doc deprecate norm and add seealso to linalg.norm (#57986)

heitorschueroff committed 3 years ago

0da54218

Add tools/ script to limit circleci to a set of jobs (#58001)

driazati committed 3 years ago

e385aa86

Add `pad_sequence` as a native function (#57868)

eqy committed 3 years ago

18edb77a

make ddp logging api to be private (#57999)

zhaojuanmao committed 3 years ago

ac44569b

[DataLoader] Remove redundant len >= 0 (#57951)

ejguan committed 3 years ago

e0539b0b

Enable cusolver potrf batched for Cholesky decomposition when cuda >= 11.3 (#57788)

xwang233 committed 3 years ago

7faac089

enable static graph training in DDP (#55248)

zhaojuanmao committed 3 years ago

ea421fb2

OpInfo: chunk (#57935)

kshitij12345 committed 3 years ago

502eb664

refactor multi_head_attention_forward (#56674)

Basil Hosmer committed 3 years ago

90f05c00

Merge branch 'master' of github.com:pytorch/pytorch into ci-all/v0dro-cusolver-lu

v0dro committed 3 years ago

2869c4b5

Add dot implementation for BFloat16 on CUDA (#57903)

imaginary-person committed 3 years ago

4fb8676c

Enable BFloat16 for `logaddexp` & `logaddexp2` on CUDA (#57908)

imaginary-person committed 3 years ago

067147ac

Enable geometric ops, exp2, expm1, rsqrt & erfc for BFloat16 on CUDA (#57913)

imaginary-person committed 3 years ago

fa318911

[CUDA graphs] Sync after replay (#57556)

mcarilli committed 3 years ago

dbedb1fa

[iOS GPU][perf][5/n] Replace std:vector with IntArrayRef and SmallVector (#57668)

xta0 committed 3 years ago

565550d8

v0dro committed 3 years ago

a900c9fc

[fbgemm] fix bug handling bias in rowwise quantization of FC (#58022)

jspark1105 committed 3 years ago

dc55ab3f

Update docs to mention CUDA support for Future (#50048)

lw committed 3 years ago

3e46d6c9

combine consecutive layes on the same device (#55973)

mrzzd committed 3 years ago

9e94921a

Merge branch 'master' of github.com:pytorch/pytorch into ci-all/v0dro-cusolver-lu

v0dro committed 3 years ago

fea486c6

make pivoted array ranged without condition

v0dro committed 3 years ago

81e5822a

Use RPC context streams to cover serde ops (#57926)

mrshenli committed 3 years ago

cf7a0e5a

fix generation of pivot tensor and removal of nan

v0dro committed 3 years ago

f677bca7

[iOS GPU][Perf][4/n] Reuse the same command buffer when copying results to CPU (#57667)

xta0 committed 3 years ago

0d564904

Deprecate chain_matmul (#57735)

lezcano committed 3 years ago

43f6deb6

Newer Older