strisunshinewentingwang/pytorch

Pull Requests Commits

ger is an alias to outer, not the other way around (#59710)

albanD committed 3 years ago

8a7c0d08

[Binary] Link whole CuDNN for CUDA-11.1 (#59802)

malfet committed 3 years ago

c2c35c01

[torch] Remove using directive from header (#59728)

enpefb committed 3 years ago

60ba4517

[After fix] Reuse constant and bump bytecode to v5 (#59722)

cccclai committed 3 years ago

e9e9291d

[torch][segment_reduce] Add support for mean reduction (cpu) (#59521)

Serhat Yilmaz committed 3 years ago

ac6b5bea

.jenkins: Ignore exit code of nvidia-smi (#59826)

seemethere committed 3 years ago

e71db0bb

[DataLoader] Add option to refine type during runtime validation for DP instance (#56066)

ejguan committed 3 years ago

e7ad82eb

[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

Update on "[DDP] Support for multiple backwards"

rohan-varma committed 3 years ago

3593da97

Update base for Update on "[DDP] Support for multiple backwards"

rohan-varma committed 3 years ago

3c65cad3

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

Newer Older