strisunshinewentingwang/pytorch

Pull Requests Commits

Remove USE_CUDA from ProcessGroupGloo (#59561)

lw committed 3 years ago

b4c35d7a

[nnc] Limit the number of inputs to a fusion group.

bertmaher committed 3 years ago

b5e83211

[nnc] Do not fuse matmul/conv2d if inputs are discontiguous. (#59754)

bertmaher committed 3 years ago

df759a3d

[ONNX] remove raw export type (#59160)

garymm committed 3 years ago

4b913552

[Static Runtime] Add schema check to several aten ops (#59603)

Hao Lu committed 3 years ago

2112074f

Disable cuDNN persistent RNN on A30 (#59830)

eqy committed 3 years ago

6eabbea4

Automated submodule update: FBGEMM (#59715)

facebook-github-bot committed 3 years ago

455afdf9

[package] doc string cleanup extravaganza (#59843)

Lilyjjo committed 3 years ago

c7890b4a

Fix torch.angle on aarch64 (#59832)

malfet committed 3 years ago

54bfd41a

[docs] Add table of contents to torch.package docs (#59842)

Meghan Lele committed 3 years ago

4025f95a

[docs] Add explanation section to torch.package docs (#59833)

Meghan Lele committed 3 years ago

0e222db0

[docs] Add "how do I" section to torch.package docs (#59503)

Meghan Lele committed 3 years ago

062dde72

[docs] Add tutorials section to torch.package docs (#59499)

Meghan Lele committed 3 years ago

6a18ca7a

[docs] Add torch.package documentation preamble (#59491)

Meghan Lele committed 3 years ago

a3db8e0a

Forward AD formulas batch 3 (#59711)

albanD committed 3 years ago

a524ee00

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

Newer Older