strisunshinewentingwang/pytorch

Pull Requests Commits

remove abi uncertainty and potential abi conflict (#94306)

jingxu10 committed 1 year ago

8b37eff6

[MPS] Fixes for Binary ops with casting issues from FP to uint8 (#94382)

kulinseth committed 1 year ago

02ca2253

Revert "[functorch] linearize (#94173)"

pytorchmergebot committed 1 year ago

e0e4f1a8

[functorch] linearize (#94173)

kshitij12345 committed 1 year ago

b6b9e1e6

Align input memory format and grad memory format for GroupNorm backward (#92668)

CaoE committed 1 year ago

81e31835

[SDPA] Adds basic correctness checks (#94274)

drisspg committed 1 year ago

81bbee7d

[Inductor] added aten.geometric_ decomp (#91672)

min-jean-cho committed 1 year ago

92f569fe

Decouple PT2 dynamic shapes from the functorch setting (#94469)

ezyang committed 1 year ago

c028fc4e

Update autocast policy list on CPU (#92527)

CaoE committed 1 year ago

c82bb287

[FSDP][optim_state_dict] Remove the dead code (#94448)

fegin committed 1 year ago

2180a0dc

[PT-D] Update torch.distributed code owners (#94362)

fegin committed 1 year ago

af5b0918

Upgrade nightly wheels to ROCm5.4.2 (#93090)

BLOrange-AMD committed 1 year ago

11f51e79

[MPS] Replace the explicit commit in View ops with adaptive commit (#94218)

razarmehr committed 1 year ago

cb715c26

[ONNX] Update CI onnx and ORT version (#94439)

titaiwangms committed 1 year ago

6d722dba

[vision hash update] update the pinned vision hash (#94455)

pytorchmergebot committed 1 year ago

03b9569d

[inductor] Fix args in sink_cat_after_pointwise (#94416)

bertmaher committed 1 year ago

bc26890b

Revert "feat(fx): `make_fx` should be aware of functions wrapped with `@fx.wrap` (#93273)"

pytorchmergebot committed 1 year ago

fe007225

[PT-D][Tensor parallelism] Add documentations for TP (#94421)

fduwjj committed 1 year ago

41e31892

[MPS] Add 2d grid sampler (#94273)

DenisVieriu97 committed 1 year ago

5b8e485a

[MPS] Fix correctness issues with Pool2D ops (#94348)

razarmehr committed 1 year ago

6c80d0a5

Revert "Set torch.backends.cudnn.enabled to false when testing accuracy (#94363)"

pytorchmergebot committed 1 year ago

ca63040d

[Executorch][Quant][BE] Refactor Choose_Qparams (#94338)

JacobSzwejbka committed 1 year ago

bb48d90b

[BE] Merge isinstance calls together (#94419)

Skylion007 committed 1 year ago

1e2d82b8

Remove duplicate CI jobs between pull and trunk (#94426)

albanD committed 1 year ago

f9cc12ee

Update xla image tag (#94377)

yeounoh committed 1 year ago

5ea6f598

[Inductor] added aten.cauchy_ decomp (#92047)

min-jean-cho committed 1 year ago

66ae3aa0

Dynamo: Support min / max over iterables (#94350)

jbschlosser committed 1 year ago

0ce95c3a

Avoid guarding on zero-ness with meta tensors. (#94399)

ezyang committed 1 year ago

53a5c8c7

Track and record hint on SymNode and use when possible (#94201)

ezyang committed 1 year ago

dc70b00d

Dynamo: Fix graph break when iterating over tensor (#94326)

jbschlosser committed 1 year ago

b5ef37b9

Newer Older