strisunshinewentingwang/pytorch

Pull Requests Commits

Provide option to save quantized data for DNNLOWP without layout optimization (#19681)

Yinghai Lu committed 5 years ago

56977db4

Fixed log_normal and geometric for CPU (#19938)

izdeby committed 5 years ago

ca57dd93

Forbid kernels from returning Scalar[] (#19811)

smessmer committed 5 years ago

c6cb32f5

Automatic generation of unittest for Glow integration

Yinghai Lu committed 5 years ago

9a81d1e6

Fix flake8. (#19832)

Mikhail Zolotukhin committed 5 years ago

3a0727e5

Fix C10_MOBILE macro for ios (#19779)

smessmer committed 5 years ago

e710f3b1

Automatic update of fbcode/onnx to f1311e74ec8a91cbf86094cd6f10157cbf00c536 (#19949)

houseroad committed 5 years ago

965c3b27

Added .bool() method

izdeby committed 5 years ago

aa6403ba

Improve performance of Int8SpatialBN (needed for DF4 quantization) (#19702)

dskhudia committed 5 years ago

d868c975

enable CopyVector for type of int32_t (#19931)

Long Jin committed 5 years ago

95ce7966

Make moving FunctionSchema efficient (#19773)

smessmer committed 5 years ago

4c678dbe

Add onnx export for floor, ceil, log2 and prim::shape

BowenBao committed 5 years ago

8e775067

Remove operator.h's dependency on function_schema.h (#19817)

zdevito committed 5 years ago

55c719b1

Add a pattern-based fusion pass. (#19596)

Mikhail Zolotukhin committed 5 years ago

2a95cf63

Fix devtoolset7 binary builds (#19919)

pjh5 committed 5 years ago

abbfb7dd

ONNX Export Min and Max ops with dim

lara-hdr committed 5 years ago

f5c2b5a2

Don't include TensorMethods.h - it's already included by Tensor.h anyway. (#19831)

Mikhail Zolotukhin committed 5 years ago

a5b8a7d4

Improve torch.utils.tensorboard docs (#19915)

orionr committed 5 years ago

62a16406

Avoid printing debug information in the test

jianyuh committed 5 years ago

ff33c8c2

Break circular dependency between Type.h, Tensor.h and TensorMethods.h (#19830)

Mikhail Zolotukhin committed 5 years ago

11464444

remove scalar to float matching (#19918)

Elias Ellison committed 5 years ago

ccf35f4b

Misc pickler improvements (#19638)

davidriazati committed 5 years ago

4294dba9

Remove self-include. (#19833)

Mikhail Zolotukhin committed 5 years ago

8933ff65

Enabled masked for a bool tensor (#19140)

izdeby committed 5 years ago

de19eeee

Add magma for CUDA 10.1 to Windows docs

peterjc123 committed 5 years ago

39b885cb

Build caffe2/fb/operators (#19688)

xw285cornell committed 5 years ago

9d0b5a1c

Finer grained consistency check in reducer (#19901)

pietern committed 5 years ago

84136002

Only call into reducer if torch.is_grad_enabled() (#19897)

pietern committed 5 years ago

5525c419

Remove trailing whitespace flake8 lint (#19828)

suo committed 5 years ago

a92e1ccd

Make find_unused_parameters in DDP default to False (#19895)

mrshenli committed 5 years ago

b695e562

Newer Older