Python: Advanced Guide to Artificial Intelligence

Logistic regression for multiclass classification

When there are more than two classes involved, the logistic regression is known multinomial logistic regression. In multinomial logistic regression, instead of sigmoid, we use softmax function that is one of the most popular functions. Softmax can be represented mathematically as follows:

Softmax function produces the probabilities for each class, and the probabilities vector adds to 1. While predicting, the class with highest softmax value becomes the output or predicted class. The loss function, as we discussed earlier, is the negative log-likelihood function -l(w) that can be minimized by ...