daily machine learning

About Me
Review

Laplace-smoothing

Published: Tue 28 April 2020

By svtter

拉普拉斯平滑，是一种用于处理0概率问题平滑方法。通过对每个分量+1来计算概率。例如，某个词语K，在不同类中的观测数量分别是0，10，20，概率为0，\(\frac{1}{3}\)，\(\frac{2}{3}\)，对这三个量进行laplace-smoothing后，结果应该是 \(\frac{1}{30+3}\), \(\frac{11}{30+3}\), \(\frac{21}{30+3}\)。

There are comments.

links

Blog site
Python.org

social

atom feed
Github
Another social link

Proudly powered by Pelican, which takes great advantage of Python.

The theme is by Smashing Magazine, thanks!