Experiments for the Course Project

This repository contains code and data for running experiments used in the course project.

Repository Structure

• datasets/ — input data used in experiments.
• algorithms.py — algorithms for approximation of mvc.
• main.py — driver code.
• stats.py — script that draws histograms and checks p-values.
• README.md — this instruction file.

How to Reproduce the Experiments

1. Install Dependencies

You need git, Python 3.13 and some required libraries. Install them in an isolated manner using:

git clone https://github.com/artem-burashnikov/greedy_mvc_benchmarks.git && \
cd greedy_mvc_benchmarks && \
python3 -m venv venv && \
source venv/bin/activate && \
pip install -r requirements.txt

3. Running experiments

You should set up your test environment first so that OS doesn't affect the results.

To run experiments execute

python3 main.py

By default, all agorithms will be tested against all datasets.

4. Veifying the results

After collectiong benchmark data you can run

python3 stats.py <bench_results.txt>

To check if benchmark data satisfies normal distribution.

License

The project is licensed under a BSD-3-Clause License.