Files
elasticsearch-js/docs/reference/api/MlCommonTokenizationConfig.md
2025-05-28 10:02:30 -05:00

662 B

MlCommonTokenizationConfig [interface-MlCommonTokenizationConfig]

Name Type Description
do_lower_case boolean Should the tokenizer lower case the text
max_sequence_length integer Maximum input sequence length for the model
span integer Tokenization spanning options. Special value of -1 indicates no spanning takes place
truncate MlTokenizationTruncate Should tokenization input be automatically truncated before sending to the model for inference
with_special_tokens boolean Is tokenization completed with special tokens