Torch.jit.script RuntimeError: Function was not annotated as having type None, but does not return along all paths:

jit
1

I met below error when doing torch.jit.script(model) on Pop!_OS with python3.8. It’s using torch 1.6.0.
The same code worked on CentOS with python3.6. But Pop!_OS with python3.8 is where I need to run the code. Looking forward for any inputs.

Traceback (most recent call last):
  File "/envs/lib/python3.8/site-packages/torch/jit/__init__.py", line 1516, in script
    return torch.jit._recursive.create_script_module(obj, torch.jit._recursive.infer_methods_to_compile)
  File "/envs/lib/python3.8/site-packages/torch/jit/_recursive.py", line 318, in create_script_module
    return create_script_module_impl(nn_module, concrete_type, stubs_fn)
  File "/envs/lib/python3.8/site-packages/torch/jit/_recursive.py", line 376, in create_script_module_impl
    create_methods_from_stubs(concrete_type, stubs)
  File "/envs/lib/python3.8/site-packages/torch/jit/_recursive.py", line 292, in create_methods_from_stubs
    concrete_type._create_methods(defs, rcbs, defaults)
RuntimeError: 
Function was not annotated as having type None, but does not return along all paths:
  File "/scripts/nn_models.py", line 47
    def forward(self, x: Tensor, hidden: Tuple[(Tensor, Tensor)]) -> Tuple[(Tensor, Tuple[(Tensor, Tensor)])]:
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~...  <--- HERE
        """Neural net forward pass. It is automatically called when use instance of LSTMnet.```
2

Could you update PyTorch to the latest release and see if you would still see the same error?
If so, could you post a minimal, executable code snippet to reproduce the error, please?

3

Thank you for your response!
This problem seems not a problem of pytorch. The problem is related to the compiling to .pyc.