原书出处
Natural Language Processing in Action, Second Edition
Now, you can insert the new token into the original document, using the match start and stop: >>> newdrug = bad_word_trans[drug] >>> if doc[start].shape_[0] == 'X': ... newdrug = newdrug.title() >>> newtext = doc[:start].text_with_ws + newdrug + " " >>> newtext += doc[stop:].text >>> newtext 'Acetaminophen costs $0.10 per tablet' Now, you have a complete pipeline, not only for detecting errors but also for replacing
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-10.html
EPUB SHA-256:0e9a35e49ca446e322e5524e53f39fb6bee5b2b23d9cda10ef052aca770b0b36