strisunshinewentingwang/pytorch

Pull Requests Commits

Do not define the macro "isnan" (#43242)

xuhdev committed 4 years ago

f4b6ef9c

Remove erroneous trailing backslashes (#43318)

xuhdev committed 4 years ago

7b520297

Implement first draft of autograd benchmark. (#40586)

albanD committed 4 years ago

c2511bdf

Autograd better error (#43308)

rakshithvasudev committed 4 years ago

0cb52cb4

Add benchmark for performance comparison (#43221)

taivu committed 4 years ago

da036250

[ONNX] Add support for operator `add` between tensor list (#41888)

BowenBao committed 4 years ago

da70976e

Extends test_unary_ufunc.py with numerics, contiguity, domain tests (#42965)

Mike Ruberry committed 4 years ago

c64594f5

Add alias torch.fix for torch.trunc to be compatible with NumPy. (#43326)

hameerabbasi committed 4 years ago

e31cd462

Bias Correction Implementation (#41845)

edmundw314 committed 4 years ago

17f9edda

Replace Conv1d with Conv2d (#42867)

taivu committed 4 years ago

665da61d

Search on system path for Vulkan headers and libraries as a last resort. (#43301)

Ashkan Aliabadi committed 4 years ago

e8139624

[quant] Make OP_LIST_TO_FUSER_METHOD public (#43286)

jerryzh168 committed 4 years ago

217ddea9

Remove proprietary notices

dreiss committed 4 years ago

844d469a

[quant][graphmode][fx] Add support for conv module (#43285)

jerryzh168 committed 4 years ago

9984d335

Reimplement per-operator selective build (#39401)

ezyang committed 4 years ago

7c50c2f7

remove empty override pretty_print (#43341)

lixinyu committed 4 years ago

e32d014f

[vulkan][ci] Vulkan tests running on linux build via swiftshader (added to docker) (#42614)

IvanKobzarev committed 4 years ago

ad8294d3

Fix backward compatibility test (#43371)

malfet committed 4 years ago

5cf85926

.circleci: Use dynamic docker image for android (#43356)

seemethere committed 4 years ago

9a1f2b36

Fix `at::native::view_as_real()` for ComplexHalf Tensors (#43279)

malfet committed 4 years ago

e10aa476

[quant][graphmode][fx][test] Add per op test for graph mode quant on fx (#43229)

jerryzh168 committed 4 years ago

b0ec3364

Update hardcoded pytorch_android_gradle_custom_build_single hash (#43340)

ezyang committed 4 years ago

2b7108a9

Make grad point to bucket buffer in DDP to save memory usage (#41954)

zhaojuanmao committed 4 years ago

97d594b9

Fix torch.hub for new zipfile format. (#42333)

Ailing Zhang committed 4 years ago

51bab087

[quant][graphmode][fx] Add graph mode quantization on fx (#43175)

jerryzh168 committed 4 years ago

dae2973f

Replace black_list with block_list (#42088)

priyanshuwustl committed 4 years ago

c89d2c6b

Minor RPC doc fixes (#43337)

mrshenli committed 4 years ago

a12fe1a2

Make TensorPipe the default backend for RPC (#43246)

mrshenli committed 4 years ago

5006d243

[ROCm] skip test_rpc in .jenkins/pytorch/test.sh (#43305)

jeffdaily committed 4 years ago

d0a6819b

vmap: Fix bug with x * 0.1 (#43218)

zou3519 committed 4 years ago

c66ca7a4

Newer Older