strisunshinewentingwang/pytorch

Pull Requests Commits

Rename tensorexpr::Value so that it can coexist with torch::jit::Value (#66467)

bertmaher committed 2 years ago

c1c985a2

[SR] Fix bug in ValueGroup (#66470)

Hao Lu committed 2 years ago

6634570a

[PyTorch][Static Runtime] Don't use vector in ProcessedNode (#65429)

swolchok committed 2 years ago

d30397d4

Cumsum Converter (#66376)

FusionPower committed 2 years ago

c6f0dde3

Use `torch.empty()` instead of `torch.tensor()` in `torch.nn.Parameter` (#66486)

cbalioglu committed 2 years ago

160946e3

Migrate USE_MAGMA config macro to ATen (#66390)

peterbell10 committed 2 years ago

30d9fd9c

remove a few unused THCTensor/Storage methods (#66555)

ngimel committed 2 years ago

e75de4f3

log_sigmoid: Use log1p for improved precision (#66441)

peterbell10 committed 2 years ago

4e1c0755

Remove native_functions.yaml dependency from Activation.cu (#64499)

ezyang committed 2 years ago

24202f7f

dropout update in autodiff (#66273)

jjsjann123 committed 2 years ago

eb8138d8

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

Newer Older