Labs from Tulane's CMPS3240 Intro to Machine Learning class. Completed independantly, by Rena Repenning, during Fall '20. Code utilizes Numpy for data set manipulation and matplotlib.pyplot for graphing.
HW 1 - My own implementation of a Perceptron learning algorithm.
HW 2 - Quantifying an approximation of training functions using VC Bound
HW 3 - Analyzing a data set with 5% noise using PLA, Pocket, and a Linear Regression. 3b Demonstrates these algorithms on a 5D data set.
HW 4 - "Demonstrating leave-one-out cross validation for nonlinear regression": using regularized regression without cross validation.
HW 5 - Identifying handwriten digits using two-dimensional features (average-intensity and symmetry); implementation of k-nearest neighbor classifier and RBF classifier algorithms.
HW 6 - Create a neural network using forward propogation