lion-ai

/

MedImageInsights

like 68

Follow

thelion.ai 48

I downloaded the weights from https://huggingface.co/lion-ai/MedImageInsights/tree/main/2024.09.27/vision_model, and got the following issue when trying to load it in Python:

>>> p = '/workspace/weights/MedImageInsights/2024.09.27/vision_model/medimageinsigt-v1.0.0.pt'
>>> a = torch.load(p)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/workspace/envs/miniconda3/envs/med/lib/python3.11/site-packages/torch/serialization.py", line 1553, in load
    raise pickle.UnpicklingError(_get_wo_message(str(e))) from None
_pickle.UnpicklingError: Weights only load failed. In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source.
Please file an issue with the following so that we can make `weights_only=True` compatible with your use case: WeightsUnpickler error: Unsupported operand 16

Check the documentation of torch.load to learn more about types accepted by default with weights_only https://pytorch.org/docs/stable/generated/torch.load.html.

I then tried specifying weights_only=False, only to get:

>>> a = torch.load(p, weights_only=False)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/workspace/envs/miniconda3/envs/med/lib/python3.11/site-packages/torch/serialization.py", line 1554, in load
    return _legacy_load(
           ^^^^^^^^^^^^^
  File "/workspace/envs/miniconda3/envs/med/lib/python3.11/site-packages/torch/serialization.py", line 1802, in _legacy_load
    magic_number = pickle_module.load(f, **pickle_load_args)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
_pickle.UnpicklingError: invalid load key, '\x10'.

Did anybody know what's going on? I've tried re-downloading the weights but it didn't help.

I was using Python 3.11.13 [GCC 11.2.0] on linux, with pytorch '2.8.0+cu128' version.