View Source tflite_beam_basic_tokenizer (tflite_beam v0.3.8)

Runs basic tokenization such as punctuation spliting, lower casing.

Related link: https://github.com/tensorflow/examples/blob/master/lite/examples/bert_qa/ios/BertQACore/Models/Tokenizers/BasicTokenizer.swift

Summary

Functions

split_by_whitespace(BinaryText)

tokenize(Text, IsCaseInsensitive)

-spec tokenize(binary() | list(), boolean()) -> [binary()].

Tokenizes a piece of text.