You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
To reproduce the bug(Bug description):
Use the pretrained weight of bert in thispage, and run pip install transformers , which will download the newest version of transformers(version 4.34.1). when trying to test the model, the Error occurs: RuntimeError: Error(s) in loading state_dict for ReferIt3DNet_transformer: Unexpected key(s) in state_dict: "language_encoder.embeddings.position_ids".
Bug explanation:
Because the pretrained weight of bert does not support the newest version of transformers(version 4.34.1) (this issue is written on 2023-Oct-24)
Solution: pip install transformers==4.30.2 to go back to an early version.
The text was updated successfully, but these errors were encountered:
To reproduce the bug(Bug description):
Use the pretrained weight of bert in thispage, and run
pip install transformers
, which will download the newest version of transformers(version 4.34.1). when trying to test the model, the Error occurs: RuntimeError: Error(s) in loading state_dict for ReferIt3DNet_transformer: Unexpected key(s) in state_dict: "language_encoder.embeddings.position_ids".Bug explanation:
Because the pretrained weight of bert does not support the newest version of transformers(version 4.34.1) (this issue is written on 2023-Oct-24)
Solution:
pip install transformers==4.30.2
to go back to an early version.The text was updated successfully, but these errors were encountered: