google-research / electra

ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
Apache License 2.0
2.33k stars 352 forks source link

about tagging task #99

Open LastRyan opened 3 years ago

LastRyan commented 3 years ago

tagging_task.py,

line 132,

if len(words_to_tokens) + len(input_ids) + 1 > self.config.max_seq_length: to if len(word_tokens) + len(input_ids) + 1 > self.config.max_seq_length: ?