Interactive maze game with neural network AI that learns to navigate labyrinths. Experience machine learning in action through gameplay!
- Intelligent Navigation:
- 🤖 Neural network-controlled character
- 🧠 Real-time decision visualization
- 🔄 On-demand training mode
- Dynamic Environments:
- 🌀 Procedurally generated mazes
- 🎚️ Adjustable complexity levels
- Game Modes:
- 👨💻 Manual player control
- 🤖 AI autonomous navigation
- 🏁 Progress tracking with scoring
- Immersive Experience: 🔊 Sound effects 🎨 Vector-based graphics 📊 Performance metrics
-
Clone repository
git clone https://github.com/Zebaro24-Stud/Neural-Network-Labyrinth.git cd Neural-Network-Labyrinth -
Create virtual environment
python -m venv venv # Windows venv\Scripts\activate # Linux/Mac source venv/bin/activate
-
Install dependencies
pip install -r requirements.txt
python main.py- Arrow Keys: Move character (Up, Down, Left, Right)
graph LR
A[Input Layer<br>4 directions] --> B[Hidden Layer<br>64 neurons]
B --> C[Output Layer<br>4 actions]
style A fill:#4CAF50,stroke:#388E3C
style B fill:#2196F3,stroke:#1976D2
style C fill:#FF9800,stroke:#F57C00
- Optimizer: Adam
- Loss Function: Categorical Crossentropy
- Activation: ReLU (Hidden), Softmax (Output)
main.py # Game entry point
GUI2D.py # Graphical interface
LabyrinthLogic.py # Maze generation & logic
NeuralNetwork.py # AI implementation
VectorGraphics.py # UI components
config.py # Game settings
sound/ # Audio assets
requirements.txt # Dependencies| Manual Mode | AI Mode |
|---|---|
![]() |
![]() |
| 👨💻 Developer | 💻 GitHub | |
|---|---|---|
| Denys Shcherbatyi | zebaro.work@gmail.com | @Zebaro24 |
| Dmitro Itsko | cloud.me.fr@gmail.com | @WhiteDh |
| Oleksandr Yakovenko | wu662542@gmail.com | @Glek-Stud |

