strisunshinewentingwang/pytorch

Pull Requests Commits

rename torch.Assert to torch._assert (#47763)

vkuzo committed 4 years ago

a8ca042e

[PyTorch] Optimize ~intrusive_ptr for the case of zero weak references (#47834)

swolchok committed 4 years ago

16d6af74

[quant] skip tests without fbgemm support (#47800)

supriyar committed 4 years ago

ed20e327

[OpBench] add _consume_op.list for processing input with type of List[Tensor] (#47890)

Yang Wang committed 4 years ago

9ee4f499

Fix some flaky tests in test_torch.py and test_nn.py (#46941)

zasdfgbnm committed 4 years ago

0652d755

CUDA BFloat16 Dropout (#45005)

zasdfgbnm committed 4 years ago

2712acbd

[quant][graphmode][fx] insert_observer_for_input_arg_of_observed_node (#47785)

jerryzh168 committed 4 years ago

1589ede8

Move eq.device to lite interpreter

cndn committed 4 years ago

dfd94687

Profiler benchmark fix (#47713)

Ilia Cherniavskii committed 4 years ago

a97c7e2e

[quant][graphmode][fx][refactor] insert_observer_for_output_of_the_node (#47784)

jerryzh168 committed 4 years ago

1afdcbfb

Support MatMul in c2_pt_converter

scambret committed 4 years ago

59e96c55

[quant][graphmode][fx][refactor] insert_observer_for_special_module (#47783)

jerryzh168 committed 4 years ago

c4ecbcdc

[ONNX] Add export of prim::dtype, prim::tolist (#46019)

KsenijaS committed 4 years ago

9fa681c5

[ONNX] Convert _len based on the first dimension length (#47538)

jiafatom committed 4 years ago

85c43c3d

[NNC] Remove all deferred expansion from Reductions (#47709)

nickgg committed 4 years ago

eab80937

Revert D24524219: Remove `balance` and `devices` parameter from Pipe.

Natalia Gimelshein committed 4 years ago

eb8331e7

[pytorch][bot] update mobile op deps (#47825)

ljk53 committed 4 years ago

4f538a2b

[pytorch] strip out warning message ifdef STRIP_ERROR_MESSAGES (#47827)

ljk53 committed 4 years ago

a376d3dd

[OpBenchMobile] Enable operator_benchmark to run the benchmark on mobile through AiBench (#47767)

Yang Wang committed 4 years ago

8ff0b6fe

Make empty c10-full (#46092)

smessmer committed 4 years ago

edf751ca

[numpy] `torch.sqrt` : promote integer inputs to float (#47293)

kshitij12345 committed 4 years ago

3649a2c1

[hotfix] fix misleadingly summary BLAS=MKL when there's no BLAS install (#47803)

Rong Rong committed 4 years ago

7391edb5

[FX] Fix submodule naming for subgraph split (#47869)

James Reed committed 4 years ago

9734c042

Added serialization of parameters for leaf modules (#47729)

gcatron committed 4 years ago

21f447ee

Remove `balance` and `devices` parameter from Pipe. (#46804)

pritamdamania committed 4 years ago

8da75763

Update, appease, and enable fail-on for shellcheck (#47786)

samestep committed 4 years ago

65d5004b

Give hash in commit messages in doc push scripts (#47694)

samestep committed 4 years ago

8304c25c

[NNC] Fix lowering of aten::pow (#47795)

nickgg committed 4 years ago

b1a4170a

Added CUDA support for complex input for torch.solve (#47045)

IvanYashchuk committed 4 years ago

149190c0

[Docs] Store Docs fixes about HashStore API (#47643)

osalpekar committed 4 years ago

275a89a7

Newer Older