Skip to content

Checkpoint loading could potentially be dangerous? #264

@EmmaRenauld

Description

@EmmaRenauld

Note. When loading checkpoint, the call torch.load(pickle_path) failed after torch 2.6, with warning:

Weights only load failed. This file can still be loaded, to do so you have two options, do those steps only if you trust the source of the checkpoint.

(1) 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.
(2) GLOBAL numpy.core.multiarray.scalar was not an allowed global by default. Please use
torch.serialization.add_safe_globals([numpy.core.multiarray.scalar]) context manager to allowlist this global if you trust this class/function.

Tried option (2), not working. Adding weights_only=False

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions