strisunshinewentingwang/pytorch

Pull Requests Commits

[pytorch] include all overloads for OSS custom build

ljk53 committed 4 years ago

a01e91e6

[quant] Use PlaceholderObserver instead of Fp16Observer and NoopObserver (#42348)

supriyar committed 4 years ago

38bf5be2

[quant][graph] Add support for FP16 dynamic quant (#42222)

supriyar committed 4 years ago

6bd46b58

[quant] Add FP16Observer for fp16 quant support (#42221)

supriyar committed 4 years ago

8c5bf102

[quant] Add saturate_to_fp16 op for FP16 quant support (#42147)

supriyar committed 4 years ago

a9eebaf6

Support RowWiseSparseAdam on GPU (#35404)

Yan Xie committed 4 years ago

bdd9ef19

[jit] make clone works for interface type (#42121)

wanchaol committed 4 years ago

a9e7e787

Revert D22812445: Update TensorPipe submodule

ezyang committed 4 years ago

352e15f1

Fix missing attribute when loading model from older version (#42242) (#42290)

sharvil committed 4 years ago

832b1659

[gloo] change ProcessGroupGlooAsyncTest to use gtest (#42313)

Brandon Lin committed 4 years ago

4c6878c9

Make resize_ use normal device dispatch (#42240)

Ailing Zhang committed 4 years ago

0adb5843

Warns when TensorIterator would resize its output (#42079)

Mike Ruberry committed 4 years ago

2f840b16

Enables torch.full bool and integer type inference (#41912)

Mike Ruberry committed 4 years ago

e54f268a

torch.where : Scalar Support (#40336)

kshitij12345 committed 4 years ago

31d41f98

Abstract cuda calls made from `torch_python` (#42251)

ziab committed 4 years ago

1c8217a7

BlackList to BlockList (#42279)

DeepakVelmurugan committed 4 years ago

fbb052c2

Modify function to takes dtype as argument

scintiller committed 4 years ago

27c22b9b

Add tests to `sigmoid_backward` and `fmod` (#42289)

scintiller committed 4 years ago

b5fcd894

Remove 4 unused variables in lp_pool_op.cc (#42329)

xuhdev committed 4 years ago

7d6c4f62

fix quantized elu benchmark (#42318)

vkuzo committed 4 years ago

153673c3

import freeze (#42319)

Elias Ellison committed 4 years ago

5ff54ff4

Let bfloat16 support promotion with other types (#41698)

xuhdev committed 4 years ago

344defc9

Add typing support to torch._six (#42232)

malfet committed 4 years ago

c489bbe1

Implementing NumPy-like function torch.signbit() (#41589)

Kiyosora committed 4 years ago

26d58503

[pytorch][ci] install nightly instead of stable libtorch for mobile CIs (#42220)

ljk53 committed 4 years ago

c35faae1

Const-correctness, variable initialization, and error checking. (#42124)

Ashkan Aliabadi committed 4 years ago

ce546328

Add missing header guards. (#42272)

Ashkan Aliabadi committed 4 years ago

d0ed1e30

Add Vulkan Test to ATen Mobile Tests. (#42123)

Ashkan Aliabadi committed 4 years ago

ee215037

Disable validation layers in non-debug builds. (#42122)

Ashkan Aliabadi committed 4 years ago

7cd92aaa

Fix minor typo in comment (#42184)

ezyang committed 4 years ago

8e3d1908

Newer Older