❓ [Question] Have anyone encounter this: RuntimeError: expected type comment but found 'eof' here

[laisimiao]

❓ Question

when I run compile command like this:

trt_ts_module = torch_tensorrt.compile(model,
                                inputs=[torch_tensorrt.Input((1, 3, 128, 128), dtype=torch.float32),
                                        torch_tensorrt.Input((1, 3, 320, 320), dtype=torch.float32)],
                                enabled_precisions = {torch.float, torch.half})

I encounter this error:

  File "/opt/conda/lib/python3.8/site-packages/torch/jit/frontend.py", line 310, in build_def
    type_comment_decl = torch._C.parse_type_comment(type_line)
RuntimeError: expected type comment but found 'eof' here:
#     # type: (List[Tensor], Tensor) -> Tensor

What you have already tried

No other attempts.

Environment

Build information about Torch-TensorRT can be found by turning on debug messages

• PyTorch Version (e.g., 1.0): 1.11.0a0+b6df043
• CPU Architecture: amd64
• OS (e.g., Linux): ubuntu18.04
• How you installed PyTorch (conda, pip, libtorch, source): pip
• Build command you used (if compiling from source):
• Are you using local sources or building from archives:
• Python version: 3.8
• CUDA version: 11.5
• GPU models and configuration: GTX 1660ti
• Any other relevant information:

Additional context

I just use docker recommended by tutorial at https://catalog.ngc.nvidia.com/orgs/nvidia/containers/pytorch

[narendasan]

Can you provide more context? Was your model already in TorchScript or are you letting TorchTRT handle that for you?

[github-actions]

This issue has not seen activity for 90 days, Remove stale label or comment or this will be closed in 10 days