Python implementations of the exercises presented by Andrew Ng in "Machine Learning" class on Coursera.

You may find more details in Russian at timofey.pro

Week 1: Linear Regression with One Variable

1. Week1: Solution using Pandas and Numpy
2. Week1: Solution using scikit-learn

Week2: Linear Regression with multiple variables

1. Week2: Solution using Numpy and feature normalization
2. Week2: Same solution using Normal Equation

Week3: Logistic Regression

1. Week3: Solution using Gradient Descent
2. Week3: Same solution using Scikit-Learn
3. Week3: Microchips solution (adding additional higher order polynomial features)

Week4: One-vs-All + Neural Networks

Week 5: Backpropagation