Python Implementation of Iterative Reweighted Least Square of Logistic Regression

This is the python implementation of Logistic Regression models from scratch. The model has been tested on the "IRIS" dataset. Only class "Virginica" and "Versicolor" has been used. Newton Raphson methodolgy was used to find parameters. The result was tested with 10-fold cross validation(also implemted from scratch).

Explanation

• Bishop notations were used.

Loss & ROC plot

Loss Plot

ROC plot