renarepenning / machinelearninglabs

Labs from Tulane's CMPS3240 Intro to machine learning class, taken during Fall '20

Intro to Machine Learning Labs

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