strisunshinewentingwang/pytorch

Pull Requests Commits

[XROS][ML] System specific adjustments for UTs to work. (#65245)

Karol Kosik committed 2 years ago

eb3b9fe7

GELU acc OP (#65957)

FusionPower committed 2 years ago

363ccb25

Compile without -Wno-unused-variable (#65954)

malfet committed 2 years ago

a6280ab6

Fix shape inference dim_type for Clip, Mean, Div (#65996)

qxy11 committed 2 years ago

10f62942

model_dump: Fix memory computation when both constants and data tensors are present (#66006)

dreiss committed 2 years ago

e1d963e8

model_dump: Add a helper to produce html with a single call (#66005)

dreiss committed 2 years ago

23caeb3f

Upload test failures to RDS (#65873)

driazati committed 2 years ago

d9a95e66

[fx2trt]add support for torch.tile (#66016)

842974287 committed 2 years ago

f85d7422

Forward fix type hint for DataLoader (#66001)

ejguan committed 2 years ago

060e41ea

Revert D30634700: [pytorch][PR] Fix typo in tensor docs

suo committed 2 years ago

ad889d0b

[fx-acc] add acc_op optimization flags and decorator (#65928)

alexbeloi committed 2 years ago

7d220079

Fix typo in tensor docs (#64160)

Akshit Khurana committed 2 years ago

d9374737

Re-generate workflows (#66027)

driazati committed 2 years ago

8e869528

Remove usage of GitHub's artifact store in linux jobs (#65875)

driazati committed 2 years ago

894d296b

Fix typo in name of LayerNormBackwardCUDAKernel (#66000)

bertmaher committed 2 years ago

6e8ffd19

[PyTorch][Static Runtime] Fast path for contiguous to_copy (#65499)

swolchok committed 2 years ago

ffede499

[PyTorch] Try removing Android strtod implementation (#65713)

swolchok committed 2 years ago

7b10a76e

[PyTorch] Fix many Tuple::elements() callsites (#64065)

swolchok committed 2 years ago

176d3c6f

[fx2trt]fix slice tensor converter (#65960)

842974287 committed 2 years ago

f14e5e63

[PyTorch][easy] Use copy-and-move instead of copy-and-swap in IValue::operator= (#65826)

swolchok committed 2 years ago

21eebc9f

[fx][const_fold] Refactor to use base split module to simplify, and correctly handle non-single-Tensor outputs (#65933)

jfix71 committed 2 years ago

592481a5

[iOS][CI] Update dev certs (#66004)

xta0 committed 2 years ago

34682377

Revert D31317680: [pytorch][PR] Avoid saving self for`softmax` and `log_softmax`

suo committed 2 years ago

ccf8d48f

suppress mypy error (#66003)

suo committed 2 years ago

21da6ae9

Revert "Port `sort` kernel to structured kernels. (#62391)" (#65876)

bdhirsh committed 2 years ago

eac218db

Avoid saving self for`softmax` and `log_softmax` (#65242)

soulitzer committed 2 years ago

5f7cadc7

Fix internal assert failure for torch.all and torch.any with requires_grad=True (#65714)

IvanYashchuk committed 2 years ago

383c0a38

Make autograd codegen for differentiable outputs safer to use (#65823)

IvanYashchuk committed 2 years ago

53c0d91d

[nnc] Add BufHandle.store to python API (#65213)

bertmaher committed 2 years ago

bff8d8fd

[nnc] Add call_with_numel interface for fast CUDA calls (#65213)

bertmaher committed 2 years ago

8cf047af

Newer Older