chunked/nltk_chunking/tasks_sequence_classification/chunk_25.txt

from transformers import DataCollatorWithPadding
data_collator = DataCollatorWithPadding(tokenizer=tokenizer)
</pt>
<tf>py
from transformers import DataCollatorWithPadding
data_collator = DataCollatorWithPadding(tokenizer=tokenizer, return_tensors="tf")

Evaluate
Including a metric during training is often helpful for evaluating your model's performance.