Jon Chuang 2cf0cf8137 [dynamo / DDP] - lazily compile submodules - to propagate real tensor strides to backend compiler (#114154) ... Fixes https://github.com/pytorch/pytorch/issues/113812, https://github.com/pytorch/pytorch/issues/102591, Probably fixes: https://github.com/pytorch/pytorch/issues/113740, https://github.com/pytorch/pytorch/issues/113786, https://github.com/pytorch/pytorch/issues/113788 Pull Request resolved: https://github.com/pytorch/pytorch/pull/114154 Approved by: https://github.com/wconstab, https://github.com/yf225		2023-12-06 18:50:14 +00:00
..
__init__.py
common.py
cudagraphs.py
debugging.py
distributed.py	[dynamo / DDP] - lazily compile submodules - to propagate real tensor strides to backend compiler (#114154)	2023-12-06 18:50:14 +00:00
inductor.py
onnxrt.py	[ONNX] switch from onnxscript-preview to onnxscript (#109139)	2023-09-18 22:24:47 +00:00
registry.py
tensorrt.py
torchxla.py	Delete a bunch of type-ignores (#113990)	2023-11-18 02:48:38 +00:00
tvm.py