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
Token indices sequence length is longer than the specified maximum sequence length for this model (585 > 512). Running this sequence through the model will result in indexing errors
modules.py WARNING: 28-Dec-24 17:01:58 - Truncating sequence from 585 to 512
...
File "/home/kbg/.miniconda3/envs/udtube/lib/python3.12/site-packages/transformers/models/bert/modeling_bert.py", line 1073, in forward
buffered_token_type_ids_expanded = buffered_token_type_ids.expand(batch_size, seq_length)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: The expanded size of the tensor (585) must match the existing size (512) at non-singleton dimension 1. Target sizes: [32, 585]. Tensor sizes: [1, 512]
Somehow the "target size" on dimension 1 has not been properly truncated despite the warnings that it has been.
It should be easy to simulate this error by feeding it a very long sentence (e.g., over 512 pieces on English BERT).
The text was updated successfully, but these errors were encountered:
Somehow the "target size" on dimension 1 has not been properly truncated despite the warnings that it has been.
It should be easy to simulate this error by feeding it a very long sentence (e.g., over 512 pieces on English BERT).
The text was updated successfully, but these errors were encountered: