Closed
Description
I am trying to lower GPT2 torch model to linalg in this link:
nod-ai/SHARK-Studio#341
nod-ai/SHARK-Studio#337
but get the following bug.
File "torch-mlir/build/tools/torch-mlir/python_packages/torch_mlir/torch_mlir/__init__.py", line 233, in compile
mb.import_module(scripted._c, class_annotator, import_options)
raise Exception(f"""
Exception:
PyTorch TorchScript module -> torch-mlir Object Graph IR import failed with:
### Importer C++ Exception:
Unsupported import tensor type: (1,1,.,.) .....
[ CPUByteType{1,1,1024,1024} ]
I am planning to add a datatype of Byte just like this patch #1169?
Any suggestion @silvasean?
Metadata
Assignees
Labels
No labels
Activity