strisunshinewentingwang/pytorch

Pull Requests Commits

[kineto] Update Kineto (severity levels in Logger.h)

ilia-cher committed 3 years ago

3e7d3b06

[reland] .github: Add Windows GPU workflow (#58782) (#59752)

seemethere committed 3 years ago

e2c784d9

.github: Ensure cleaner windows workspace (#59742)

seemethere committed 3 years ago

54cc477e

fx quant: remove some dead code in observer insertion (redo) (#59799)

vkuzo committed 3 years ago

0099c25b

[WIP] Add slow gradcheck build for the ci/slow-gradcheck label (#59020)

albanD committed 3 years ago

fb620a27

Fix Error when run python setup.py install again on Windows (#59689)

mszhanyi committed 3 years ago

cc32dcad

Fixing and enabling tests that check fake_quant matches quant+dequant (#59095)

HDCharles committed 3 years ago

1fc3576d

[fix] torch.{lin, log}space(): properly examine passed dtype (#53685)

kshitij12345 committed 3 years ago

c9026090

Split slow gradcheck periodic CI job so that it does not time out (#59736)

soulitzer committed 3 years ago

9bcef86d

displays graph node's info (#59679)

Edvard Ghazaryan committed 3 years ago

f2406240

[skip ci] export_slow_tests.py - Add option to ignore small differences (#59759)

janeyx99 committed 3 years ago

7af9252e

Link ATEN tests with OpenMP runtime (#59733)

malfet committed 3 years ago

51d954e8

[PyTorch ] Thread parallel bmm across batch dim (#59596)

kimishpatel committed 3 years ago

4f79270b

[Pytorch benchmark] Add BMM benchmark (#59595)

kimishpatel committed 3 years ago

3176f166

Added doc for torch.einsum sublist format (#57038)

heitorschueroff committed 3 years ago

58412740

Update TensorPipe submodule (#59789)

lw committed 3 years ago

5e3e5047

Automated submodule update: tensorpipe (#59374)

facebook-github-bot committed 3 years ago

96651458

[fx ir] Support lists and dicts in FX IR GraphDrawer (#58775)

khabinov committed 3 years ago

0d7d316d

Add query and synchronize to c10::Stream (#59560)

lw committed 3 years ago

e7cccc23

Support EnumerableShardingSpec in ShardedTensor. (#59061)

pritamdamania committed 3 years ago

f1112096

[C10d] Support subgroups (#59111)

Yi Wang committed 3 years ago

48ea7c80

[c10d] Use TORCH_CHECK for monitored barrier error (#59667)

rohan-varma committed 3 years ago

fc0582ee

Bump the bytecode reading version kMaxSupportedBytecodeVersion to 6 (#59714)

iseeyuan committed 3 years ago

12b9e99e

[OSS CI][iOS] Use LibTorch-Lite.h for nightly builds (#59762)

xta0 committed 3 years ago

3c6ae6f1

ci: Add skipIfOnGHA util (#59748)

seemethere committed 3 years ago

a62f6b6d

Add USE_WHOLE_CUDNN option (#59744)

malfet committed 3 years ago

1ea5c19c

add channels last support for AvgPool2d on CPU (#58725)

mingfeima committed 3 years ago

bb19dc14

Revert D29007258: Revert D28926135: [pytorch][PR] Refactor Foreach Tests: Unary Functions

ngimel committed 3 years ago

52b2ed65

Update Kineto to fix fd leak (#59755)

malfet committed 3 years ago

827e00c9

Comment on tests reliance on ZeRO's partitioning algo (#59713)

awgu committed 3 years ago

a4e0368c

Older