strisunshinewentingwang/pytorch

Pull Requests Commits

Revert D20289209: Support RowWiseSparseAdam on GPU

ezyang committed 4 years ago

d927d58c

Revert D20497453: [pytorch][PR] Makes floor_divide a method, adds sparse floor division

Mike Ruberry committed 4 years ago

a1eaaea2

Do not throw from CUDAContext destructor (#34756)

malfet committed 4 years ago

a3de3594

Makes floor_divide a method, adds sparse floor division (#34552)

Mike Ruberry committed 4 years ago

b7129050

[caffe2] open source 2/4-bit SLS operators (#34903)

jspark1105 committed 4 years ago

bcbdba45

[C++ API Parity] LBFGS optimizer step() update and added closure to the Optimizer step() function (#34564)

anjali411 committed 4 years ago

d7e4a379

Updating submodules

svcscm committed 4 years ago

df20f5b3

[torchbind] Add more comprehensive docscrings (#34906)

James Reed committed 4 years ago

130e7207

[torchbind] Improve IValue custom class API and remove most Capsule stuff (#34848)

James Reed committed 4 years ago

09a7788a

Support using self as the destination in rpc.remote for builtin operators (#34931)

mrshenli committed 4 years ago

c4fdba32

[JIT] Make RPC RRef Owner WorkerInfo.name available to TorchScript (#34896)

xush6528 committed 4 years ago

b5edf329

Revert D20480546: adds quantized implementation of hard sigmoid

mrshenli committed 4 years ago

95f1cb34

[rpc] handle exceptions in ProcessGroupAgent::enqueueRecv (#34413)

rohan-varma committed 4 years ago

ff3d205e

[quant][graphmode][refactor] Change QParamMap to QParamVector (#34314)

jerryzh168 committed 4 years ago

1c8e0865

fix barrier in jit test (#34901)

zhaojuanmao committed 4 years ago

4bd3e9b4

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

Newer Older