strisunshinewentingwang/pytorch

Pull Requests Commits

JIT: guard DifferentiableGraph node (#49433)

t-vi committed 3 years ago

ea087e2d

[fix] torch.cat: Don't resize out if it is already of the correct size. (#49937)

kshitij12345 committed 3 years ago

36ddb002

Change CMake config to enable universal binary for Mac (#50243)

janeyx99 committed 3 years ago

c2d37cd9

Support scripting classmethod called with object instances (#49967)

ppwwyyxx committed 3 years ago

49bb0a30

Optimize Vulkan command buffer submission rate. (#49112)

Ashkan Aliabadi committed 3 years ago

1c12cbea

add type annotations to torch.nn.modules.fold (#49479)

guilhermeleobas committed 3 years ago

aa18d174

Unused exception variables (#50181)

alexhenrie committed 3 years ago

2c4b6ec4

Fix MKL builds on Ubuntu (#50212)

antocuni committed 3 years ago

8f31621f

Revert D25717504: Clean up some type annotations in test/jit

heitorschueroff committed 3 years ago

1bb7d8ff

Apply clang-format to rpc cpp files (#50236)

mrshenli committed 3 years ago

f9f758e3

Dump state when hitting ambiguous_autogradother_kernel. (#50246)

Ailing Zhang committed 3 years ago

0bb341da

Convert string => raw strings so char classes can be represented in Python regex (#50239)

Thomas Zhang committed 3 years ago

d78b638a

[numpy] torch.{all/any} : output dtype is always bool (#47878)

kshitij12345 committed 3 years ago

5d45140d

Clean up some type annotations in test/jit (#50158)

r-barnes committed 3 years ago

a4f30d48

[onnx] Do not deref nullptr in scalar type analysis (#50237)

malfet committed 3 years ago

81778e28

Improve torch.linalg.qr (#50046)

antocuni committed 3 years ago

b5ab0a7f

Stop using c10::scalar_to_tensor in float_power. (#50105)

gchanan committed 3 years ago

88bd69b4

add type annotations to torch.nn.quantized.modules.conv (#49702)

guilhermeleobas committed 3 years ago

55919a47

Fix persistent_workers + pin_memory (#48543)

ssnl committed 3 years ago

54ce171f

Add tensor.view(dtype) (#47951)

zasdfgbnm committed 3 years ago

d00acebd

Implement torch.linalg.svd (#45562)

antocuni committed 3 years ago

5c5abd59

Update autograd related comments (#50166)

albanD committed 3 years ago

006cfebf

[numpy] torch.exp: promote integer inputs to float (#50093)

kshitij12345 committed 3 years ago

9f832c8d

Autograd engine, only enqueue task when it is fully initialized (#50164)

albanD committed 3 years ago

fc2ead09

Revert D25687465: [PyTorch] Devirtualize TensorImpl::dim() with macro

luciang committed 3 years ago

c215ffb6

Address clang-tidy warnings in ProcessGroupNCCL (#50131)

rohan-varma committed 3 years ago

294b7867

Disable cuDNN persistent RNN on sm_86 devices (#49534)

xwang233 committed 3 years ago

5a63c452

[PyTorch] Change representation of SizesAndStrides (#47508)

swolchok committed 3 years ago

b73c0185

[PyTorch] Introduce packed SizesAndStrides abstraction (#47507)

swolchok committed 3 years ago

882ddb2f

Completely remove FutureMessage type (#50029)

mrshenli committed 3 years ago

c480eebf

Newer Older