https://github.com/abdellatif-laghjaj/caartpole-reinforcement-learning

https://github.com/abdellatif-laghjaj/caartpole-reinforcement-learning

Last synced: 8 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/abdellatif-laghjaj/caartpole-reinforcement-learning
• Owner: abdellatif-laghjaj
• Created: 2025-01-03T10:41:33.000Z (10 months ago)
• Default Branch: main
• Last Pushed: 2025-01-04T13:57:14.000Z (10 months ago)
• Last Synced: 2025-02-22T17:45:15.371Z (8 months ago)
• Language: Jupyter Notebook
• Size: 38.1 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# CartPole Reinforcement Learning

Deep Q-Learning implementation for the CartPole balancing problem using PyGame and TensorFlow.

# Screenshot



## Features

- DQN with TensorFlow

- Real-time PyGame visualization

- Experience replay

- Live training progress

## Setup

```bash

pip install -r requirements.txt

python cartpole.py

```

## Implementation

- Neural Network: 2 hidden layers (64 units)

- Episodes: 100

- Batch Size: 32

- Learning Rate: 0.001

- Memory: 2000 experiences