Skip to content

README.md

Latest commit

 

History

History
43 lines (33 loc) · 3.1 KB

README.md

File metadata and controls

43 lines (33 loc) · 3.1 KB

🧬 Genetic Algorithm: Mathematical Function Optimization

This project implements a Genetic Algorithm (GA) from scratch in Python, designed to find the maximum point of a given mathematical function (defaulting to a quadratic function). The code is modular, highly configurable, and automatically generates both detailed execution logs and graphical representations of the population's evolution.

evolution_plot

📌 Description

By default, the algorithm searches for the maximum of a quadratic function on a defined interval. To achieve this, it uses binary encoding for chromosomes and applies classic genetic operators across multiple generations.

Key Features:

  • Configurability: The config.py file can be edited to simulate the evolution of populations under varied parameters.
  • Binary Encoding: Transforms real-valued numbers into binary strings with high precision (configurable).
  • Roulette Wheel Selection: Includes a Fitness Scaling implementation to correctly handle negative fitness values.
  • Single-point Crossover: Exchanges genetic material between selected pairs of individuals.
  • Uniform Mutation: Flips bits (0 to 1 and vice versa) based on a defined probability.
  • Elitism: Guarantees that the absolute best individual (the Elite) passes into the next generation completely unaltered.
  • Logging System: Automatically saves detailed metrics of each generation into a text file (evolution.txt).
  • Graphical Visualization: Generates a plot (evolution_plot.png) at the end of the execution to visualize algorithm convergence (Max Fitness vs. Average Fitness).

🛠️ Project Structure

  • main.py - The entry point of the application. It manages the main generational loop and handles file writing/plotting.
  • config.py - The configuration file containing all the algorithm's hyperparameters (population size, probabilities, precision, etc.).
  • ga/ - The core package containing the genetic logic:
    • individual.py - Defines the Individual class, encoding/decoding methods, and fitness calculation.
    • genetic_algorithm.py - The main steps of the GA (selection, crossover, mutation, parent replacement).
    • operators.py - The mathematical logic behind the genetic operators.
    • population_generator.py - Handles the generation of the initial random population.
  • utils/ - Package for optional utility functions:
    • logger.py - Dedicated functions for formatting and displaying population information (probabilities, intervals, mutations).
    • plotter.py - Functions for generating and saving performance charts using Matplotlib.
  • data/ - Directory designated for storing charts and the text file containing data of one simulation.

⚙️ Requirements & Installation

This project requires Python 3.x. The external libraries required are:

  • numpy (for random number generation and vectorized operations)
  • matplotlib (for generating evolution plots)

You can install the requirements using pip:

pip install numpy matplotlib