fix: unpickling-error-pytorch-2.6#72
Open
NebulaMao wants to merge 1 commit into
Open
Conversation
NebulaMao
force-pushed
the
fix/unpickling-error-pytorch-2.6
branch
from
September 8, 2025 17:31
9e206ec to
969c14e
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
issue: #71
This pull request improves the robustness of the YOLO model loading process in
modules/subcore.pyby adding compatibility for different PyTorch versions. The main change is the introduction of a fallback mechanism to ensure that the model loads correctly, regardless of whether the newersafe_globalsfeature is available.Model loading compatibility improvements:
load_yolofunction to first attempt loading the model using PyTorch 2.6+'ssafe_globalscontext, and if that fails (due to missing features or incompatible types), it falls back to the previous loading method. If both attempts fail, the error is logged and re-raised for better debugging.