strisunshinewentingwang/pytorch

Pull Requests Commits

Autograd: Delay warnings until the end of backward execution (#66235)

peterbell10 committed 2 years ago

5f45927d

[GHA] Hardcode doc build target to `master` (#66567)

malfet committed 2 years ago

42328090

[jit] Refcounting spot fixes in alias_analysis (#66295)

swolchok committed 2 years ago

0aab34c2

[jit] Add MutableTypePtrHelper::mapTypeToBorrowedAliasTypeSet (#65344)

swolchok committed 2 years ago

97672826

[jit] Implement one-element MemoryDAG::mayContainAlias more efficiently (#65178)

swolchok committed 2 years ago

75d98fa0

[fx2trt][code quality] Add type annotation and docstring to utils functions in acc_ops_converters.py (#66496)

842974287 committed 2 years ago

9e8281fd

[Static Runtime] Clone test does not use uninitialized memory (#66557)

Mike Iovine committed 2 years ago

37db650c

fix lint (#66572)

suo committed 2 years ago

82986a17

Disable .numpy() and .tolist() for tensor subclasses subclasses and fix .tolist() for conjugated and negated tensors (#66082)

anjali411 committed 2 years ago

a82fcd35

[qnnpack] Remove usage of conv_param_t in deconv-run.cc (#66465)

John Shen committed 2 years ago

675ba6cd

Add OpInfo for torch.bucketize (#65821)

saketh-are committed 2 years ago

86cf22cb

Handle shared memory cases in MathBithFallback (#63602)

anjali411 committed 2 years ago

035310c5

Make empty* and *_like factory functions respect tensor subclasses (#65677)

lezcano committed 2 years ago

c04bcde2

Skip `interactive_embedded_interpreter.cpp` for clang-tidy (#66569)

malfet committed 2 years ago

b792a778

.github: Enable onnx tests (#66513)

seemethere committed 2 years ago

09b90612

Make ContainerHash compatible with const& types (#66497)

wconstab committed 2 years ago

f48f20e1

add a note on numerical accuracy (#65947)

ngimel committed 2 years ago

fdd9f49c

Use interactive_embedded_interpreter to dynamicly loading various third-party libraries (#66512)

shunting314 committed 2 years ago

a453ebc8

[vulkan] Remove the persistent resource pool (#66478)

SS-JIA committed 2 years ago

a8815d55

[vulkan] Release GPU resources when vTensor::View is destroyed (#66477)

SS-JIA committed 2 years ago

cebaf21c

[FX] Fix cases when we should not fuse due to more than one users of intermediate node (#66472)

Yinghai Lu committed 2 years ago

5e34ac6c

[oss/ci] skip all dataloader tests with asan (#66561)

suo committed 2 years ago

9d13ae45

Add no-input-grad-needed cases to test_grid_sample (#66071)

to-mi committed 2 years ago

713e025c

Compute input gradient only if required (CUDA) (#66070)

to-mi committed 2 years ago

8a40bb62

Compute input gradient only if required (CPU) (#66069)

to-mi committed 2 years ago

f8d98b5a

Add output_mask (#66068)

to-mi committed 2 years ago

84385c40

fix type error in hipify_python.py (#66164)

arpancodes committed 2 years ago

6401658b

Add softplus support to autodiff (#63942)

jjsjann123 committed 2 years ago

d8594889

Add tensor.{adjoint(),H,mT,mH} methods and properties (#64179)

lezcano committed 2 years ago

82a216c4

[Bootcamp][Pytorch]Add testing for complex parameters in Adagrad optimizer (#66501)

Christopher Gray Howard committed 2 years ago

87df043f

Newer Older