-
Notifications
You must be signed in to change notification settings - Fork 5.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Loading LORA weights with load_lora_weights
do not respect HF_HUB_OFFLINE
/ local_file_only
#6089
Comments
Maybe it's better to be explicit here so, I would prefer the first option. |
Agree on raising an explicit error while the problem is not solved but it would be good to be able to load lora weights while being offline, no? I think it just requires to list the local files ( And I'll open an issue in |
Hmm, so when |
I created huggingface/huggingface_hub#1899 in |
@Wauplin thanks for your inputs. I will take a closer look soon and open a PR asking for your review. |
First reported by @rvorias in huggingface/huggingface_hub#1893 (comment).
TL;DR: Loading lora weights on a pipeline makes a call to the Hub even when
HF_HUB_OFFLINE=1
is set as environment variable (orlocal_file_only=True
is passed as parameter), which is not expected.This seems to be due to the
_best_guess_weight_name
method that try to guess which file to load when the LoRA filename is not provided. It makes a call tomodel_info
to check which files exist on the Hub repo.Suggestions:
The text was updated successfully, but these errors were encountered: