原书出处
Designing Large Language Model Applications
Loss = max(d(a, p) – d(a, n) + margin, 0) where d is a distance measure, typically Euclidean distance. The margin is a hyperparameter that represents the distance by which the negative example should be farther away from the anchor than the positive example. When using Euclidean distance as the distance measure, I suggest a margin of 5, but make sure to tune it if
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch11.html
EPUB SHA-256:1b4d2347fec2f949fcddfbaf4fad597a1ae20887a096cfae64c59a619b4781bf