The aim of this project is to learn to use the concepts of machine learning presented in the lectures and practiced in the labs on a real-world dataset, in the context of a recent popular machine learning challenge - finding the Higgs boson -, using original data from CERN~\cite{higgs}. For any further information on this machine learning challenge and its dataset, feel free to consult the original paper. The ID of our best submission is 164381
This project is organized as follows :
- the repository src that includes:
- proj1_helpers.py which contains the helper functions provided by the Professor
- implementations.py which contains all the other functions coded by ourselves used to create this machine learning system
- the file run.py which produces exactly the same .csv predictions we used in our best submission to the competition system
- project1.ipynb which contains the additional code that shows several results and figures produced for the analysis of our ML system
- the file brr-submission.pdf which is the report of our group brr that provides a full explanation of our ML system and our findings.
Just make sure to have the libraries mentioned below installed on your environment before running the cells in the jupyter notebook. In order to reproduce our predictions, please download the dataset available on the competition arena in a repository named data at the same level of repository.
In this project we used a few libraries other than NumPy for vizualisation purposes :
- matplotlib.pyplot
- seaborn
Luca Bataillard - 282152 Julian Blackwell - 289803 Changling Li - 282440