FYS-STK4155 - Applied Data Analysis and Machine Learning

This repository contains projects for Machine Learning Course taken at the University of Oslo.

Project 1:

In this project, I performed Linear Regression and its variations (Ridge, Lasso) on the Franke function and terrain data.

All files, codes and the pdf of the report can be found inside "p1" folder.

Project 2:

In this project, I tried to predict the probability of default from credit card data using Logistic Regression and Feed Forward Neural Network developed from scratch.

All files, codes and the pdf of the report can be found inside "p2" folder.

Project 3:

In this project, we performed the analysis of the microbial ecology data set comprised of 72 observations and 15000 predictors using Random Forests, XGboost, Siamese and Feed-Forward Neural Networks, where the combination of scikit-learn, TensorFlow/Keras and the codes developed during first two projects were used for performance comparisons.

Note: Third project was a team effort performed by Laurent Fontaine, Maziar Kosarifar, Maksym Brilenkov and Knut Hauge Engvik. The files in this repository was/is a backup in case of "something went wrong". If you want to check the actual report, please see this repository instead: https://github.com/maziark/FYS-STK-project3