strisunshinewentingwang/pytorch

Pull Requests Commits

Ignore shape inference exception from Caffe2 ATen fallback (#90408)

Thiago Crepaldi committed 1 year ago

1d4e8723

[aot autograd] merge all outputs of funtionalization analysis into single metadata (#95991)

bdhirsh committed 1 year ago

98ece750

aot autograd: handle detach() and no_grad() mutations on input (#95980)

bdhirsh committed 1 year ago

29b216ac

[inductor] do not handle `int` in `placeholder` (#96230)

nkaretnikov committed 1 year ago

bb650b34

aot autograd: dont allow symint outputs to get tangents in the bw graph (#96219)

bdhirsh committed 1 year ago

f96bd528

Revert "Fix hooks handling for unpickled nnmodule (#96224)"

pytorchmergebot committed 1 year ago

6bbae862

Hooking backward for QNNPACK (#94432)

kwanghoon-meta committed 1 year ago

a1d7014c

[FSDP][optim_state_dict] Fix a memory leakage in optim_state_dict (#96263)

fegin committed 1 year ago

92edac72

[MPS] Adding xfaillist with all categories of failures. (#96176)

kulinseth committed 1 year ago

2bb022e9

[static-runtime] fix one forwarding usage (#96271)

tenpercent committed 1 year ago

b90a9c7d

Revert "[Dynamo] Support torch.{cuda/cpu}.amp.autocast (#95416)"

pytorchmergebot committed 1 year ago

3ce1e15c

`dl_open_guard` should restore flag even after exception (#96231)

malfet committed 1 year ago

941ff109

Fix hooks handling for unpickled nnmodule (#96224)

wconstab committed 1 year ago

8ca264ef

[Quant] Add lowering for pixel_unshuffle/narrow (#96160)

jiaxuzhu92 committed 1 year ago

08fb13db

[1/n] Add verifier for EXIR Aten dialect (#94783)

qihqi committed 1 year ago

9e3f1736

Use GH cache for sccache on GH mac runners (#96142)

clee2000 committed 1 year ago

3a427527

Context manager to push/pop frame summaries (#96054)

voznesenskym committed 1 year ago

d7db5b05

[vision hash update] update the pinned vision hash (#96243)

pytorchmergebot committed 1 year ago

bb8645ac

[CI] Avoid calling torch.use_deterministic_algorithms for some models (#96245)

desertfire committed 1 year ago

664381b2

[ET] Add RuntimeContext to ET Aten mode (#96084)

kirklandsign committed 1 year ago

93ff71ec

[Dynamo] Support torch.{cuda/cpu}.amp.autocast (#95416)

yanboliang committed 1 year ago

c88aa336

[Dynamo] Guard name should be valid Python identifier (#96174)

yanboliang committed 1 year ago

b8f7bd59

Fix validate_input_col for nn.Module or Callable (#96213)

ejguan committed 1 year ago

738cc5e6

[PyTorch][easy] Don't call IValue::type twice in Pickler::endTypeTag (#96214)

swolchok committed 1 year ago

fdd7e76b

[FSDP] Speed up first iter order check (part 2) (#96220)

awgu committed 1 year ago

3623cfb7

Add torch.empty_like() to documented list of supported nested tensor ops (#96211)

jbschlosser committed 1 year ago

7324aef9

Fix gumbel cdf (#91698)

vfonov committed 1 year ago

b0b5f3c6

Properly show buck target to run (#96089)

mlazos committed 1 year ago

203890e1

[ONNX] Fix expand_as (#95962)

guyang3532 committed 1 year ago

79d49c60

[ONNX] Skip doctest `torch.onnx._internal.fx` if ImportError (#95686)

BowenBao committed 1 year ago

bdb076ab

Newer Older