13x005-ai-project's Introduction

13X005 AI Project - Logistic Regression & Naive Bayes

End of semester AI Project on Logistic Regression and Naive Bayes using the famous "Iris" dataset.

Assignement Pdf:

regressionLogistique.pdf

For this project, a website for the presentation's slides was made:

https://unige-ai.vercel.app

DataSet Used:

Todo

Link to trello used to manage todos: https://trello.com/b/xhglaB3g/13x005-ai-project

Preview:

Building & Runing

Poetry was used to simplify the project & dependencies setup, i.e. avoid problems related to python/package versions as well as the "It works on my machine" problem.

Dependencies:
This project manages depency with poetry (installable with ./setup_poetry 1).
To install the required dependencies, simply run poetry install. Note that the makefile should do that automatically if you don't already have a poetry.lock file.

If for some reason you don't want to use poetry, you can see a list of dependencies from here,
Run: To run it just use make.
Test: To run the tests use make test or to test the effiency (f1-score) of the models use make test_model as some tests can be quite verbose.

Editing the report

You can directly edit the markdown version in report/report.md and use make report to convert it from markdown to latex and from latex pdf. (pandoc and pdflatex are required for this to work. Pandoc should be installed by default on most linux distributions.)
The report follows the LaTeX template defined in preamble_ai_project.sty, which looks like this: pdf-report
The citations are in the file report/references.bib and can be called with \cite{citation-key}.

Submission

To build a zip archive containing the relevant files for submission, run make package and an archive named ai-project-group3.zip will get created at root of the project.