strisunshinewentingwang/pytorch

Pull Requests Commits

Wrap indirect indexing on CUDA (#105055)

lezcano committed 1 year ago

2b6249e2

Make ExportedProgram valid tracing callable (#107657)

tugsbayasgalan committed 1 year ago

c81c217a

[ONNX] Add huggingface models into CI tests (#107247)

titaiwangms committed 1 year ago

400c4de5

inductor: also check index_exp when select tiling var (#106765)

XiaobingSuper committed 1 year ago

610f64d7

Enable mypy check in torch/_inductor/config.py (#107448)

FFFrog committed 1 year ago

4a40e275

[ONNX] Exclude FXSymbolicTracer from _assert_fake_tensor_mode (#107712)

titaiwangms committed 1 year ago

d0f8ee45

Fix torch.compile with FakeTensor that has SymInt sizes (#107662)

ydwu4 committed 1 year ago

31b04457

Enable Mypy Check in torch/_inductor/virtualized.py (#107127)

FFFrog committed 1 year ago

83517c8d

[MPS] Fix `torch.std` for negative dimentions (#107754)

malfet committed 1 year ago

4cc05c41

[vision hash update] update the pinned vision hash (#107757)

pytorchupdatebot committed 1 year ago

17675cb1

Fix the use of head_branch in filter-test-configs action (#107753)

huydhn committed 1 year ago

09c642bf

Fix torch.compile FunctionalTensor inputs for higherOrderOps (#107604)

ydwu4 committed 1 year ago

cbcd5510

[ez] Minor refactors (#107656)

tugsbayasgalan committed 1 year ago

fc380a2b

Add _native_batch_norm_legit_no_training to core IR (#107732)

kirklandsign committed 1 year ago

d395088d

MatMul heuristics for aarch64 (#107167)

yd2102 committed 1 year ago

d34cf147

Dispatch take_along_axis to gather (#107711)

lezcano committed 1 year ago

fada0527

[dynamo] np.sort(complex) is not implemented (#107710)

lezcano committed 1 year ago

62113a23

Support negative indices in ndarray.__getitem__ (#107688)

lezcano committed 1 year ago

2fc82831

Add a flag that allows breaking on NumPy ops (#107687)

lezcano committed 1 year ago

db39a81e

Revert "[ATen] Update pre-compiled header (#106915)"

pytorchmergebot committed 1 year ago

e573abec

[BE] reorganize opt disables in dynamo for clarity (#107709)

janeyx99 committed 1 year ago

874d1b18

fallback to cpu_kernel for VSX (#98511)

cdeepali committed 1 year ago

0c4fa022

Revert "[inductor] Adjust dynamic SMEM limit when above default in AOT (#107601)"

pytorchmergebot committed 1 year ago

42897e81

[Mypy] move inductor to exclude list (#107741)

eellison committed 1 year ago

68c941d2

[BE]: Update ruff to 0.285 (#107519)

Skylion007 committed 1 year ago

660e8060

Support graphs which return get_attr nodes directly as output (#107610)

tarun292 committed 1 year ago

e8278d60

[dtensor] update some comments (#107608)

wanchaol committed 1 year ago

979e706f

[dtensor] fix requires_grad in distribute_tensor (#107606)

wanchaol committed 1 year ago

945fa7e8

Add Early Link To Intro Doc (#107344)

Tincaniam committed 1 year ago

8367cf12

[MPS][BE] Refactor atan2_out_mps (#107334)

malfet committed 1 year ago

b115da83

Newer Older