Project for the course of Combinatorial Decision Making and Optimization - Module 2 @UniBo.
To run the code, create a virtual environment using the provided requirements file and run python3 src/main.py
.
The project is about gradient-free optimization. The map contained in this project was created by merging smaller maps downloaded from Emilia Romagna's official altimetry repository. The maps were concatenated and a mean filter was applied on the final matrix to obtain a smoother mountain.
The following agents/optimization methods were implemented:
- Nelder-Mead simplex algorithm
- Particle Swarm Optimization
- Backtracking Line Search
The final visualization/animation was created using Plotly.