HTRU2-Pulsar-Detection

Project relative to the analysis of the HTRU2 dataset through the use of Machine Learning algorithms. Machine Learning and Pattern Recognition course at Politecnico di Torino, A.Y. 2021/2022

Content of the repository

• report.pdf is the report showing the results and the choices made during the model analysis.
• main.py is the main that must be executed in order to obtain the results shown in the report.
• func_lib is a folder containing a collection of Machine Learning algorithms and auxiliary libraries.
• project_plots is a folder containing all the plots generated by the execution of main.py.

Known issues in main.py (as highlighted by the professor)

• wrong application of the score calibration in both evaluation phase and validation phase
• other minor issues that I can't recall
• missing hyperparameter analysis during evaluation (also in the report)

No ML libraries or ML toolboxes have been used in this project.