May 2019
Intermediate to advanced
162 pages
4h 24m
English
Mathematically, logistic regression is very similar to linear regression. The inner product of our parameters and X represent the log odds of the membership of a class, which is simply the natural log of the probabilities over 1 minus the probabilities:

What we really want are the probabilities of the class membership. We can back out of the log odds and determine the probabilities using the sigmoid or logistic function.
Read now
Unlock full access