This repository contains programs using classical Machine Learning algorithms to Artificial Intelligence implemented from scratch and Solving traveling-salesman problem (TSP) using an goal-based AI agent.
This project is about finding a solution to the traveling-salesman problem (TSP) using a so called goal-based AI agent. The goal is to find a cycle (a roundtrip) which visits every city once, while traveling the minimal possible distance.
A search algorithm called Simulated Annealing search has been used, which is a algorithms from the family of local search algorithms.Simulated annealing is a probabilistic technique for approximating the global optimum of a given function. Specifically, it is a metaheuristic to approximate global optimization in a large search space for an optimization problem. It is often used when the search space is discrete
For this project Qatar and Djibouti cites loaction have been used. Qater data is in DATA/qa194.tsp directory. Djibouti data is in DATA/dj38.tsp directory.
- Google Colab:
- Copy Simulated_annealing.ipynb file in your google drive
- Copy data in data folder in your google drive
- Open Simulated_annealing.ipynb in your google drive
- change data path in the code
- .py file:
- Open Simulated_annealing.py in your pyhton code editor
- Change data path in python code
- Run the code
- Fully observable
- Single agent
- Stochastic
- Episodic
- Static
- Discrete
- Known
- Project.pdf
- In this file, all details about the project have been explained.
- Report.pdf
- Report file for the project. All results of implemented python codes with plots are visible here.
- Simulated_annealing.ipynb
- Simulated Annealing google colab code
- Simulated_annealing.py
- Simulated Annealing python code
Unfortunately "Project.pdf" and "Report.docx" are only available in "Persian". If you seriously need a translation, contact me.
