Error compiling the trained model weights

Running aimnet jitcompile /home/joshua/Downloads/SPICE_amino_acid_ligands_v1_model_100.pt /home/joshua/Downloads/SPICE_amino_acid_ligands_v1_model_100.jpt results in:

Traceback (most recent call last):
  File "/home/joshua/mambaforge/envs/aimnet2_py311/bin/aimnet", line 8, in <module>
    sys.exit(cli())
             ^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/click/core.py", line 1157, in __call__
    return self.main(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/click/core.py", line 1078, in main
    rv = self.invoke(ctx)
         ^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/click/core.py", line 1688, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/click/core.py", line 1434, in invoke
    return ctx.invoke(self.callback, **ctx.params)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/click/core.py", line 783, in invoke
    return __callback(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/aimnet/train/pt2jpt.py", line 59, in jitcompile
    model: nn.Module = build_module(model)
                       ^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/aimnet/config.py", line 124, in build_module
    d[k] = get_init_module(v['class'], args=v.get(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/aimnet/config.py", line 42, in get_init_module
    return get_module(name)(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/aimnet/modules.py", line 106, in __init__
    shifts.weight = shifts.weight.to(dtype)
    ^^^^^^^^^^^^^
  File "/home/joshua/mambaforge/envs/aimnet2_py311/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1729, in __setattr__
    raise TypeError(f"cannot assign '{torch.typename(value)}' as parameter '{name}' "
TypeError: cannot assign 'torch.DoubleTensor' as parameter 'weight' (torch.nn.Parameter or None expected)

Any idea what could be the issue here?

zubatyuk / aimnet2

Error compiling the trained model weights #5