strisunshinewentingwang/pytorch

Pull Requests Commits

Make checkInputs more robust (#34838)

Krovatkin committed 4 years ago

74a28ff1

Reduce memory overhead of categorical.sample (#34900)

neerajprad committed 4 years ago

e43c2d59

Fix dist autograd context Example block format (#34921)

mrshenli committed 4 years ago

85c51a8c

Fix example block format in Distributed Optimizer API doc (#34919)

mrshenli committed 4 years ago

f05abd12

Fix example format in Distributed Autograd doc (#34914)

mrshenli committed 4 years ago

e87db8a7

Minor fixes for RPC API docs (#34890)

mrshenli committed 4 years ago

552f9d3a

Update descriptions for transmitting CUDA tensors (#34888)

mrshenli committed 4 years ago

3c48aadd

Removing experimental tag in for RPC and adding experimental tag for RPC+TorchScript (#34887)

mrshenli committed 4 years ago

800bdcf0

Adding warnings for async Tensor serialization in remote and rpc_async (#34885)

mrshenli committed 4 years ago

6446ccce

Add a warning for RRef serialization (#34884)

mrshenli committed 4 years ago

0d857d55

Append multiple arguments to list of flags as multiple items (#34899)

malfet committed 4 years ago

f87cd83d

[quant][graphmode] Quantization pattern for aten::linear (#33854)

jerryzh168 committed 4 years ago

841f7600

[RPC] Avoid polluting Python root logger on importing "torch" module (#34871)

xush6528 committed 4 years ago

71f02a48

adds quantized implementation of hard sigmoid (#34607)

vkuzo committed 4 years ago

58c5b6d3

Format register_ditributed_ops.cpp (#34922)

xush6528 committed 4 years ago

97757dca

SNIFAE Template Constructors of IValue (#34647) (#34843)

terhuhf committed 4 years ago

0216c76e

Support RowWiseSparseAdam on GPU (#34341)

Yan Xie committed 4 years ago

959a7138

[ROCm] Fix for std::isnan regression in ROCm (#34664)

rohithkrn committed 4 years ago

72e3d66f

.circleci: Remove should_run_job, no longer needed (#34326)

seemethere committed 4 years ago

b227ea95

Turn on exact_dtype by default on test_optim.py (#34825)

great-way committed 4 years ago

5857a125

Eliminate guards through max_pool ops. (#34512)

Owen Anderson committed 4 years ago

a4224886

Add types argument to __torch_function__ (#34303)

hameerabbasi committed 4 years ago

6b701de1

setup.py: Add numpy as required for install_requires (#34510)

seemethere committed 4 years ago

275f5c80

Add back cudaHostRegister to cudart API. (#34665)

ezyang committed 4 years ago

940e678d

Implement channels last upsample2d/3d forward pass kernel. (#34597)

kimishpatel committed 4 years ago

7a3cf67f

move emulation libraries to contrib (#34861)

Hector Yuen committed 4 years ago

3ad7dfa2

Fix CMake Dev warning in caffe2/CMakeLists.txt (#34886)

malfet committed 4 years ago

cfab65d9

Revert D20461609: [caffe2] open source 2/4-bit SLS operators

ezyang committed 4 years ago

3e68d0c5

[TensorExpr] Pull changes from bertmaher/pytorch_fusion. (#34842)

Mikhail Zolotukhin committed 4 years ago

95833a49

[RPC] Use qualified name str directly in RPC torch script code path (#34733)

xush6528 committed 4 years ago

ecd7c0f8

Newer Older