strisunshinewentingwang/pytorch

Pull Requests Commits

Fix for incorrect usage of logging in torch/distributed/distributed_c10d.py (#51739)

szmigacz committed 3 years ago

075bbe0d

[Gradient Compression] Add a minimum compression rate threshold for PowerSGD communication hook (#52541)

Seung-Jae Bang committed 3 years ago

2d75346c

[PyTorch Mobile] Allow loading of all extra files using the extra_file argument (#52635)

dhruvbird committed 3 years ago

755c60bf

docs: add link to Tensor.share_memory_ in Module.share_memory (#52561)

Jeff Yang committed 3 years ago

0c455332

Add default args for CUDA stream and events

nikithamalgifb committed 3 years ago

ea0725d5

Allow Transformer accept output result that is not Proxy (#52473)

842974287 committed 3 years ago

238b0bbb

Fix hipify_python (#52709)

malfet committed 3 years ago

75f7b220

Modernize for-loops (#52330)

r-barnes committed 3 years ago

26419815

replace type().backend() with device() (#52558)

cyyever committed 3 years ago

caa377f5

[DataLoader] TransformsIterDataPipe (#52604)

ejguan committed 3 years ago

b534466f

Fix wrong TORCH_CHECK usages (#52670)

malfet committed 3 years ago

cabb1e7a

add OneDNN pooling backward (#49454)

XiaobingSuper committed 3 years ago

420fc42e

Remove unused variable (#52652)

r-barnes committed 3 years ago

df30cb78

Move cuda9 and cuda11.2 CI jobs to a scheduled workflow (#52693)

janeyx99 committed 3 years ago

d5ed5756

[fx] Separate globals assignment from code generation (#51974)

suo committed 3 years ago

ecf3ca00

[FX acc]Store shape and dtype in serialized output node args (#52462)

842974287 committed 3 years ago

1cddb27f

[caffe2] add a Python test for SaveOp chunking

simpkins committed 3 years ago

e2afb269

Pass child error to parent in distributed tests. (#52632)

pritamdamania committed 3 years ago

1c63cb2c

Fake Quantization support for f16 and f32 (#52612)

HDCharles committed 3 years ago

e3a805b9

Ignore user annotated ignored attributes (#52367)

tugsbayasgalan committed 3 years ago

e658d7c3

Revert D26598115: [pytorch][PR] Update XNNPACK

H-Huang committed 3 years ago

2680ff77

Add OpInfo entries for i0 and logical_not (#51956)

kshitij12345 committed 3 years ago

49b59e34

Fix performance of CUDA trilinear interpolate backward (#52351)

xwang233 committed 3 years ago

dc6fab44

Update XNNPACK (#52645)

malfet committed 3 years ago

3721962c

[TensorExpr] Properly handle ExternalCalls in LoadStore analysis and Inliner. (#52628)

Mikhail Zolotukhin committed 3 years ago

64847c7f

[TensorExpr] Inlining: allow inlining into Load exprs. (#52627)

Mikhail Zolotukhin committed 3 years ago

b63a1e31

Use `int8_t` instead of `char` in [load|store]_scalar` (#52616)

malfet committed 3 years ago

67794b14

[TensorPipe] Update [Cpu|Cuda]Buffer fwd declarations (#52600)

malfet committed 3 years ago

7ecc1b60

[caffe2] Delete some unused fields from TensorProto (#52521)

simpkins committed 3 years ago

fa856818

docs: add fractional_max_pool in nn.functional (#52557)

Jeff Yang committed 3 years ago

f111ec48

Newer Older