strisunshinewentingwang/pytorch

Pull Requests Commits

Add fast path of qmean/qstd for quantized CPU (#70172)

Xia-Weiwen committed 2 years ago

c1fa9fdf

Refactor lift_subgraph_as_module as a fx.passes.util function (#80292)

SherlockNoMad committed 2 years ago

ac5a9478

[CI] imporve ios simulator test (#80459)

linbinyu committed 2 years ago

da61ec2a

Interpreter for decomposing aten -> prims (#79989)

davidberard98 committed 2 years ago

00f65181

Add ComplexDouble scalar creation bindings to nvFuser's Python API (#80522)

IvanYashchuk committed 2 years ago

92e1710d

Add integer support to scatter_reduce (#80324)

pytorchmergebot committed 2 years ago

d7847ed2

functionalization: fix _unsafe_view debug asserts (#80526)

bdhirsh committed 2 years ago

dc7d3fd4

[DataLoader] Close open in DataPipe streams on best effort basis (#78952)

VitalyFedyunin committed 2 years ago

331c0c18

[jiterator] De-template launch_jitted_reduce_kernel (#80138)

peterbell10 committed 2 years ago

2f146f1d

[primTorch] support one tensor and two scalars in _prims.where (#80146)

Aidyn-A committed 2 years ago

74fb6ee4

[jiterator] Reduce templating in jitted_gpu_kernel_impl (#80103)

peterbell10 committed 2 years ago

66f66fac

Revert "Add objective-c language support in CMake (#80432)"

pytorchmergebot committed 2 years ago

e3599b03

Increase atol for test_noncontiguous_samples_nn_functional_conv_transpose3d_cuda_float32 (#80518)

saketh-are committed 2 years ago

725de4fb

Revert "Extract setting up ec2 linux and checkout to be composite actions (#80462)"

pytorchmergebot committed 2 years ago

de0150d8

[GH1] Relanding #80064 to erase double messaging as it was overwritten by mistake (#80550)

janeyx99 committed 2 years ago

655fc51f

Extract setting up ec2 linux and checkout to be composite actions (#80462)

ZainRizvi committed 2 years ago

fe8bfef8

Add integer support for gpuAtomicMin/Max/Mul (#80320)

mikaylagawarecki committed 2 years ago

a3430106

[ONNX] Fix potentially unbound variables (#79789)

justinchuby committed 2 years ago

d1d2687d

[ci] temporarily disable ROCm distributed tests (#80530)

suo committed 2 years ago

35d97f7b

Revert "Add __all__ for torch.distributed and fx modules (#80460)"

pytorchmergebot committed 2 years ago

58532256

Add _reduce_scatter_base to ProcessGroupWrapper. (#79633)

Rodrigo Kumpera committed 2 years ago

08795f9a

torch.special.scaled_modified_bessel_k0 (#78900)

0x00b1 committed 2 years ago

63ef2a03

[primtorch] add reference for clamp_min/clamp_max (#79821)

jjsjann123 committed 2 years ago

c28315ea

Added support for `expand` in LazyTensor shape inference (#77830)

miladm committed 2 years ago

0922cc02

[torchdynamo hash update] update the pinned torchdynamo hash (#80485)

pytorchmergebot committed 2 years ago

853247e5

Clone tensor to write in ShardedTensor checkpoint (#79400)

zhuzilin committed 2 years ago

3d9cef8c

Don't error if _warned_capturable_if_run_uncaptured not set (#80345)

ezyang committed 2 years ago

57f001f3

Add __all__ for torch.distributed and fx modules (#80460)

anjali411 committed 2 years ago

5d40c3d5

[MPS] Add logical ops (#80216)

qqaatw committed 2 years ago

5943aaa0

Add objective-c language support in CMake (#80432)

qqaatw committed 2 years ago

182870f4

Newer Older