Deep Learning with PyTorch Quick Start Guide

Logistic regression

A simple logistic regression model does not look a great deal different from the model for linear regression. The following is a typical class definition for a logistic model:

Notice that we still use a linear function when we initialize the model class. However, for logistic regression, we need an activation function. Here, this is applied when forward is called. As usual, we instantiate the model into our model variable.

Next, we set the criterion and optimizer:

We still use stochastic gradient descent; however, we need ...