CSE 575 Statistical Machine Learning projects.

• Project 1 involves using an images average brightness and variance to predict if the image is a 1 or a 0 from MNIST dataset. This project applies density estimation using Naive Bayes Classification.
• Project 2 is an implimentation of K-Means clustering. Strategy 1 randomly picked k points to be the initial cluster. Strategy 2 uses an initial point as a cluster, then for the next k-1 clusters, finds them by taking the farthest average distance. Addtional functions were added to visualize where the centroids are located by plotting a graph of the data.
• Project 3 is an implementation of a simple Convolutional Neural Network (CNN) to classify some images. My assignment was to complete the evaluation code to compute the accuracy and loss of the network, given the data and the labels.