strisunshinewentingwang/pytorch

Pull Requests Commits

record_stream() for shifted view tensors (#27371)

sublee committed 4 years ago

c1c176d9

.gitignore for the docs folder

z-a-f committed 4 years ago

6e59fb6a

Fix DDP incompatibility issue with nn.MultiheadAttention. (#26826)

Guanheng Zhang committed 4 years ago

eb932003

Replace references to _DataLoaderIter with _BaseDataLoaderIter (#27105)

ngoldbaum committed 4 years ago

f522bde1

Regenerate aten_op.h when native_functions.yaml changes. (#27253)

ngoldbaum committed 4 years ago

d5712482

change input shape to reduce variation (#27548)

mingzhe09088 committed 4 years ago

31a6ff46

Move RPC API to torch.distributed.rpc

pietern committed 4 years ago

b4ce922b

Move internal functions to torch.distributed.rpc

pietern committed 4 years ago

a6d26ce1

Move RPC backend registry to torch.distributed.rpc

pietern committed 4 years ago

14f1629c

Remove torch.distributed.rpc function (#27287)

pietern committed 4 years ago

1fd14c58

Rename variables and add comments (#27286)

pietern committed 4 years ago

48a571b2

Remove shebang from non-executable files in torch.distributed

pietern committed 4 years ago

f597926f

Fix pybind11 warnings in python_rpc_handler.cpp (#27284)

pietern committed 4 years ago

c7429188

Emergency split CUDA libtorch build/test into separate job (#26859)

ezyang committed 4 years ago

0d22f3b1

fix documentation for add_hparams (#27521)

Alexander Melnikov committed 4 years ago

660264e1

Add C++ torch::nn::CosineEmbeddingLoss (#27345)

jon-tow committed 4 years ago

3b5d40c3

Use orig source range in Node::print

James Reed committed 4 years ago

e63bfb78

Updating submodules

svcscm committed 4 years ago

e2143fde

Set existing attributes under recursive script (#27514)

davidriazati committed 4 years ago

725810f4

Stops common_utils.py from setting the default tensor type (to torch.DoubleTensor) (#27444)

Mike Ruberry committed 4 years ago

7f183a97

Fixed typos and grammatical errors (#27465)

iam-abbas committed 4 years ago

16ece1c9

Revert "Make static dispatch turn off variable before entering the kernel. (#26908)" (#27283)

ezyang committed 4 years ago

6e0312a9

docstring only formatting changes: quantize.py, fake_quantize.py, observer.py

gottbrath committed 4 years ago

a96b003b

Exponential decay of the weight of task loss (#27508)

Swati Rallapalli committed 4 years ago

e63addff

Roll master to 1.4.0 (#27374)

ezyang committed 4 years ago

2c51e065

Revert D17159707: [pytorch][PR] [ONNX] Fixed Select symbolic to export slice when index = negative one

houseroad committed 4 years ago

34662f77

Updating submodules

svcscm committed 4 years ago

1b5df374

Fix broken name mangling

James Reed committed 4 years ago

84e2dc69

#include <stdexcept> into flat_hash_map.h (#27478)

pbelevich committed 4 years ago

23f2fb0a

Ensure NCCL error handling code is disabled for NCCL versions < 2.4 (#27124)

pritamdamania committed 4 years ago

24242e86

Newer Older