The goal of this implementation is to use gradient descent to find the optimal values of θ that minimize the cost function.
The implementation follows standard practices for logistic regression with regularization, and the formulas used in each method are derived from the logistic regression theory.