strisunshinewentingwang/pytorch

Pull Requests Commits

Port CUDA implementation of expm1 to ATen (#26598)

vishwakftw committed 4 years ago

aaf30cdf

Use Caffe2's implementation of grouped depthwise 3x3 convolutions (#26556)

Ashkan Aliabadi committed 4 years ago

729f8425

Refactor checked_tensor_unwrap to take DeviceType instead of Backend (#26290)

ezyang committed 4 years ago

1cae5195

quantized_tensor tests (#26784)

jerryzh168 committed 4 years ago

b0bb5e33

Lets generic tests use multiple devices (#26594)

Mike Ruberry committed 4 years ago

25cd3c6b

autodiff changes to enable profiling

Krovatkin committed 4 years ago

db5791d5

move more functions to InsertObserversHelper (#26773)

jerryzh168 committed 4 years ago

0cb10d7e

Updating submodules

svcscm committed 4 years ago

f0e507cb

Upgrade sleef to v3.4.0. (#26749)

xuhdev committed 4 years ago

ee6cdb57

Hub improvements (#26723)

Ailing Zhang committed 4 years ago

0f1fbc0e

Revert D17549623: Add some missing constructors to IValue.

ezyang committed 4 years ago

61dd485b

Updating submodules

svcscm committed 4 years ago

f094afe4

Add some missing constructors to IValue. (#26718)

ezyang committed 4 years ago

c5b57aa5

Remove unnecessary include from TensorBody (#26360)

ezyang committed 4 years ago

037cfce7

Implement multiple dispatch in boxed c10 dispatcher (#26118)

ezyang committed 4 years ago

52614f5f

Delete backwards compatibility Backend overload for registerOp (#25914)

ezyang committed 4 years ago

b56ad744

Named tensor support for logsumexp, mode, kthvalue, median, min, max (#26563)

zou3519 committed 4 years ago

33467597

Expose a torch.result_type and simplify tensor iterator

nairbv committed 4 years ago

002c2501

Support Negative Axis in Size in ONNX (#26436)

lara-hdr committed 4 years ago

5001ec42

Update ONNX Export for Interpolate in Opset 11 (#26778)

lara-hdr committed 4 years ago

d396c733

Named tensor support for: atan2, output_nr, detach{_}, requires_grad_ (#26543)

zou3519 committed 4 years ago

60343a82

Revert D17458232: Fake quantization enhancements for QAT/PTQ support

zou3519 committed 4 years ago

be93d30e

Fake quantization enhancements for QAT/PTQ support (#26420)

raghuramank100 committed 4 years ago

e2c3d7e5

Add <cinttypes> include to resolve PRIu32 macro (#26745)

pietern committed 4 years ago

a395c311

Handle DeQuantStub() for QAT (#26518)

raghuramank100 committed 4 years ago

bc4519dc

Improve error message in IR parser when accessing undefined variable.

Mikhail Zolotukhin committed 4 years ago

99496388

rename caffe2::mobile_threadpool to caffe2::mobile_pthreadpool

ljk53 committed 4 years ago

6d0b0045

Add comments for multidim tensor factory limitations, and rename ListInitTensor for better clarity (#26756)

Will Feng committed 4 years ago

d4dc844e

use new fbgemm PackedDepthWiseConvMatrix without template parameter (#26760)

jspark1105 committed 4 years ago

e54a9e1b

Revert D17559660: [fix] quantized_tensor tests

ezyang committed 4 years ago

0c0b4b63

Newer Older