原书出处
Natural Language Processing in Action, Second Edition
5.3 Skiing down the error slope The goal of training neural networks is to minimize a loss function by finding the best parameters (weights) for your model. At each step of the optimization loop, your algorithm finds the steepest way down the slope. Keep in mind, this error slope is not the error for just one example from your dataset. It is minimizing the cost
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