strisunshinewentingwang/pytorch

Pull Requests Commits

[Futures] Allow setErrorIfNeeded arg to have type FutureError (#39113)

osalpekar committed 4 years ago

a47e2d44

Restore thrust path for 1d tensors cumulative ops (#39180)

ngimel committed 4 years ago

f1170898

Revert D21781515: [Onnxifi] Generic way of passing output shape/type hints

smessmer committed 4 years ago

e286cb5e

[Onnxifi] Generic way of passing output shape/type hints (#39229)

Yinghai Lu committed 4 years ago

7b07208d

Use @skipIfNoFBGEMM instead of direct check (#39068)

malfet committed 4 years ago

c5def603

Fix return types of Windows API functions in __init__.py (#39334)

peterjc123 committed 4 years ago

e6d86036

[Futures] Added markCompletedIfNeeded API (#39080)

osalpekar committed 4 years ago

295a23f4

Check illegal output dtype for torch.{min, max} (#38850)

zasdfgbnm committed 4 years ago

48e66859

Make Optimizer.state_dict() nondeterministic (#37347)

aphedges committed 4 years ago

a3c87c49

Adding sparse Lp regularization operator to Caffe2 (#38574)

Jamie King committed 4 years ago

7f1a96d4

Made sure torchscript compiles in optimized mode (#38888)

benoitsteiner committed 4 years ago

6d3e4aa0

Automated submodule update: FBGEMM (#39322)

facebook-github-bot committed 4 years ago

f76e05a2

Don't generate DeviceGuard for CPU wrapping code. (#38806)

ezyang committed 4 years ago

cffa0bee

Remove supports_named_tensor from codegen entirely. (#38739)

ezyang committed 4 years ago

2b6a48e9

[Profiler x RPC] Enable RPC Server Global Profiler (#38847)

xush6528 committed 4 years ago

45baf0e1

Reland Refactor c10::complex and cleanup c10::Scalar (#39306)

zasdfgbnm committed 4 years ago

bdaa7849

Test PyTorch using python-3.8 + GCC-9 on Bionic (Reland) (#39121)

malfet committed 4 years ago

39d03725

Kill CPPTypeToScalarType. It's now subsumed by CPPTypeAndStdComplexToScalarType. (#39263)

gchanan committed 4 years ago

78244f81

Avoid defining bogus CPPTypeAndStdComplexToScalarType<void> by using some decltype tricks. (#39261)

gchanan committed 4 years ago

ddf6d494

[nvFuser] add torch.jit.fuser context manager (#38993)

jjsjann123 committed 4 years ago

07518e12

`verbose` unused in `torch.backends.cudnn` (#39228)

guol-fnst committed 4 years ago

42b2dee6

fake_quantize: respect device affinity (#39031)

vkuzo committed 4 years ago

c193bd41

Revert D21374247: Use TensorMethods.cpp

ezyang committed 4 years ago

2fe0fc26

Fix missing code in 'Installing C++ distribution of Pytorch' (#39237)

ShawnZhong committed 4 years ago

1943a2c3

Division by zero crashes for fmod operator(#32699) (#38919)

guol-fnst committed 4 years ago

7773a45c

DOC: remove java documentation (#38920)

mattip committed 4 years ago

dc4fd040

Migrate AT_DISPATCH_FLOATING_AND_COMPLEX_TYPES_AND2 to c10::complex (#39285)

zasdfgbnm committed 4 years ago

a5664510

Add dynamic_cast asserts to CPU Loops. (#39258)

gchanan committed 4 years ago

aa5afbdb

Make needs_dynamic_casting multiple-complex-type aware. (#39255)

gchanan committed 4 years ago

9b05b1ba

Loops: Separate out dynamic_casting concerns from complex overloads. (#39254)

gchanan committed 4 years ago

f9edbda7

Newer Older