https://github.com/bdr-pro/pathfindingtensorflowrl
showcases a pathfinding visualization using A* algorithm and Q-Learning model to find paths through a generated maze.
https://github.com/bdr-pro/pathfindingtensorflowrl
tensorflow
Last synced: 2 months ago
JSON representation
showcases a pathfinding visualization using A* algorithm and Q-Learning model to find paths through a generated maze.
- Host: GitHub
- URL: https://github.com/bdr-pro/pathfindingtensorflowrl
- Owner: BDR-Pro
- Created: 2024-04-05T17:51:16.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-04-05T22:08:12.000Z (about 2 years ago)
- Last Synced: 2025-01-21T18:37:20.864Z (over 1 year ago)
- Topics: tensorflow
- Language: Python
- Homepage:
- Size: 372 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Pathfinding with A* and Q-Learning
This project showcases a pathfinding visualization using A* algorithm and Q-Learning model to find paths through a generated maze. The comparison between these two methods provides insights into their efficiency and applicability in solving pathfinding problems. Utilizing Pygame for visualization and TensorFlow for the Q-Learning model, this project offers an engaging way to understand and analyze pathfinding algorithms and machine learning in action.
.png)
## Features
- **Maze Generation**: Randomly generates a maze for pathfinding.
- **A* Algorithm Implementation**: Utilizes the A* algorithm to find the shortest path from start to goal.
- **Q-Learning Model**: Employs a Q-Learning model developed with TensorFlow to learn and find paths through the maze.
- **Visualization**: Uses Pygame for real-time visualization of the maze, paths found by A* and the Q-Learning model, start and goal positions.
- **GPU Acceleration**: Checks for CUDA compatibility to leverage GPU acceleration for the Q-Learning model training and inference, enhancing performance.
- **Threaded Execution**: Runs pathfinding operations in separate threads to keep the UI responsive and provide real-time updates.
## Installation
Before you start, ensure you have Python 3.7+ and pip installed on your system. Clone this repository or download the code.
To install the required packages, run:
```bash
pip install pygame tensorflow keras numpy
```
## Usage
To start the application, navigate to the project directory and run:
```bash
python pathfinding.py
```
The Pygame window will open, displaying the generated maze and, eventually, the paths found by both the A* algorithm and the Q-Learning model.
## Development
This project uses:
- Python 3.7+
- TensorFlow 2.x for the Q-Learning model.
- Pygame for visualization.
- NumPy for array manipulation and operations.
## Contributions
Contributions are welcome! Whether it's bug reports, improvements, or new feature suggestions, feel free to open an issue or a pull request.
---
Dive into the fascinating world of pathfinding algorithms and reinforcement learning with this interactive visualization project!