Includes any programs I made while taking the Machine Learning course by Andrew Ng on Coursera
linear_regression.ipynb: An implementation of the linear regression algorithm taught in the course. Uses the multivariable Boston housing dataset and attempts to predict house prices.
logistic_regression.ipynb: An implementation of the logistic regression algorithm taught in the course along with weight regularization techniques. Uses a custom dataset for easy visualisation.
neural_network.ipynb: An implementation of the neural network structure taught in the course. Uses the MNIST dataset and achieves a testing accuracy of approximately 70% in under 4 epochs or 200 iterations.
support_vector_machine.ipynb: An implementation of the support vector machine algorithm taught in the course using cvxopt to solve the Quadratic Programming problem. Implements standard linear kernel and polynomial kernel for non-linearly separable cases.
k_means.ipynb: An implementation of the k-means clustering algorithm taught in the course using blobs for data and repeated trials to avoid local optima.
pca.ipynb: An implementation of the principal component analysis algorithm taught in the course. Uses blobs for data, does preprocessing step, shows visualizations of data in 3D and reduced data in 2D, and shows percentage of variance retained.
anomaly_detection.ipynb: An implementation of the anomaly detection algorithm taught in the course. Uses multivariate gaussian distributions and graphs those distributions using contour plots and 3d surface plots.
collaborative_filtering.ipynb: An implementation of the collaborative filtering algorithm taught in the course. Uses a vectorized version of the gradient descent algorithm the course shows and uses regularization to avoid overfitting.