strisunshinewentingwang/pytorch

Pull Requests Commits

Use normal dispatch to get to CUDA threshold kernels, instead of DispatchStub. (#30307)

ezyang committed 4 years ago

9740011f

Add torch.multiprocessing.create_processes

Ailing Zhang committed 4 years ago

a997f224

Add ONNX Scripting Conv Support (#30618)

lara-hdr committed 4 years ago

4d30415f

split getInvokedMethods (#30546)

jerryzh168 committed 4 years ago

89be1a22

improve .view() performance (#30554)

Natalia Gimelshein committed 4 years ago

d5c13609

Change test_invalid_names test to only test constructor of WorkerInfo (#30620)

rohan-varma committed 4 years ago

5a484245

Remove deprecated fromIvalue in RRefForkData

mrshenli committed 4 years ago

f9f54201

TestCppExtension now removes /tmp/torch_extensions folder so that it can be used by other users in a multi-user environment. (#30095)

nikitaved committed 4 years ago

b4465729

Turn off scalar_checks for _th_reciprocal. (#30436)

gchanan committed 4 years ago

8b29701a

Turn off scalar_checks for torch.clamp. (#30435)

gchanan committed 4 years ago

61798865

Raise an error for is_signed on quantized types (#30527)

nairbv committed 4 years ago

e5b947a3

Exclude undefined tensors in the result of Module::parameters() / named_paramters() / buffers() / named_buffers() (#30626)

Will Feng committed 4 years ago

18ec4632

Fix typos (#30606)

bwignall committed 4 years ago

e7fe64f6

Add the explicit per-tensor/per-channel quant info when we print the module (#30591)

jianyuh committed 4 years ago

0bebfe21

Fix serialization memory lifetime issue. (#30603)

jjlilley committed 4 years ago

4dab29a2

Fix TCPStoreTest and improve tcputils::connect() (#30354)

pritamdamania committed 4 years ago

db81e13d

Disable implicit conversion warning (#30529)

WenleiHe committed 4 years ago

9e3d1941

Add support for converting quantized AvgPool2d and Reshape operations (#30490)

supriyar committed 4 years ago

968c0d4a

Add barriers to fix flaky test_graph_for_py_nested_call and (#30624)

pritamdamania committed 4 years ago

2d0a4e42

PRAGMA missing for clang (#30351)

EscapeZero committed 4 years ago

98ab55fc

Add pickler support for Device (#30131)

davidriazati committed 4 years ago

9c02b887

Add TOC to CONTRIBUTING.md (#29671)

davidriazati committed 4 years ago

19b7d49f

Turn off scalar_checks for exp, cos, cosh, tan, atan, tanh, erf, erfc. (#30434)

gchanan committed 4 years ago

56972952

Back out "Back out "Revert D18542342: Boxed variable dispatch""

smessmer committed 4 years ago

90821230

windows build (#30556)

Mingbo Wan committed 4 years ago

3636cb03

make the order btw div and mul in adagrad update consistent (#30449)

jspark1105 committed 4 years ago

d32f261f

Use pybind11::gil_scoped_* functions instead of AutoGIL/AutoNoGIL (#30274)

ezyang committed 4 years ago

1111a6b8

Update magma to 2.5.1 for Windows and switch CUDA in CI to 9.2

peterjc123 committed 4 years ago

6deb41c8

add small input shapes to some ops (#30617)

mingzhe09088 committed 4 years ago

b68d1fc3

Fix CPU_INTEL flag error on windows (#30564)

liooil committed 4 years ago

8ee61e0b

Newer Older