原书出处
Natural Language Processing in Action, Second Edition
Momentum is an attribute of our gradient descent algorithm that allows it to “accelerate” when it’s moving in the right direction and “slow down” when it’s getting away from its target. How do we decide which values to give these two attributes? As with other hyperparameters you see in this book, you’ll need to optimize them to see what’s the most effective one for your
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-5.html
EPUB SHA-256:0e9a35e49ca446e322e5524e53f39fb6bee5b2b23d9cda10ef052aca770b0b36