原书出处
Natural Language Processing in Action, Second Edition
7.4.3 Transfer learning Another enhancement that can help your CNN models is using pretrained word embeddings, such as GloVe. And this isn’t cheating because these models have been trained in a self-supervised way, without any labels from your disaster tweets dataset. You can transfer all the learning these GloVe vectors contain from the training that Stanford gave them on all of Wikipedia and other larger
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-7.html
EPUB SHA-256:0e9a35e49ca446e322e5524e53f39fb6bee5b2b23d9cda10ef052aca770b0b36