This repository contains all the homework solutions for the Neural Networks and Machine Learning class taken at Kennesaw State University.
Basic introduction to Python 3 syntax and certain libraries.
Univariate Linear Regression. The first part included manual calculations, and then in the second part, those manual calculations were verified through some Python code.
Multiple Linear Regression and Linear Algebra. The first part included calculations or proofs of some fundamental linear algebra operations, including matrix multiplication and transpose. In the second part, a dataset containing 3 data points and 4 features was used to perform multiple linear regression using NumPy.
ML Workflow. Using the Boston Housing Dataset to improve the model's performance while following the proper workflow. Done using SciKit-Learn.
Logistic Regression Using Gradient Descent. In the first part, core concepts of logistic regression are explored and some manual calculations are performed that are essential for model training. In the second part, machine learning workflow is performed on a dataset using the gradient descent algorithm developed in class.
Learning XOR Function Using a 2-Layer Neural Network. In the first part, the parameters involved in forward and back propagation are manually calculated. Then, in the second part, the network is built using NumPy.