This repository contains the workflow I followed to acquire and apply my skills in Machine Learning during my time at Trinity College Dublin. The best way to understand the process is by going through the reports and reviewing the code that appears in the appendix of those reports.
The repository includes reports and statements, which are the problem statements that guided the creation of the reports. Additionally, it follows a typical project folder structure with src, data, and output directories, all designed to allow you to run the project by executing the main script.
I believe this approach offers a comprehensive view of how to effectively apply Machine Learning, from data preparation to result interpretation, showing a rigorous and well-documented analysis. I’ve used powerful tools to train, evaluate, and fine-tune models, always aiming to achieve the best possible results.
- Reports: Detailed documents explaining each part of the process, including comparisons between models and their performance.
- Statements: Problem statements that served as the basis for the reports and the experiments conducted.
- Folders: The repository follows a typical project structure:
- src: Contains all source code, including model implementations and functions.
- data: Holds the datasets used for training and testing the models.
- output: Stores the results and generated files, such as figures and logs.
- Main Script: The main script can be executed to run the project and reproduce the results.
By executing the main script, you can replicate the analysis, train the models, and view the results and outputs generated in the output folder.