https://github.com/hongbo-wei/lunar-lander

Implement reinforcement learning to train lunar lander to land on the moon safely. Assignment from Machine Learning Specialization by DeepLearning.AI and Standford
https://github.com/hongbo-wei/lunar-lander

ai machine-learning reinforcement-learning

Last synced: about 14 hours ago
JSON representation

Implement reinforcement learning to train lunar lander to land on the moon safely. Assignment from Machine Learning Specialization by DeepLearning.AI and Standford

• Host: GitHub
• URL: https://github.com/hongbo-wei/lunar-lander
• Owner: hongbo-wei
• Created: 2024-05-01T07:55:13.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2024-08-11T20:49:29.000Z (3 months ago)
• Last Synced: 2024-08-11T21:49:53.865Z (3 months ago)
• Topics: ai, machine-learning, reinforcement-learning
• Language: Jupyter Notebook
• Homepage: https://www.coursera.org/learn/unsupervised-learning-recommenders-reinforcement-learning
• Size: 9.76 MB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Lunar-Lander

Implement reinforcement learning to train lunar lander to land on the moon safely. Assignment from Machine Learning Specialization by DeepLearning.AI and Standford

* Write an unsupervised learning algorithm to **Land the Lunar Lander** Using Deep Q-Learning

    - The Rover was trained to land correctly on the surface, correctly between the flags as indicators after many unsuccessful attempts in learning how to do it.

    - The final landing after training the agent using appropriate parameters : 



## Instruction

Tested on *Python 3.11*

1. Create virtual environment

    ```

    python -m venv lunar_lander

    ``` 

    Activate the virtual environment on **Linux/macOS**

    ```

    source lunar_lander/bin/activate

    ```

    Activate the virtual environment on **Windows**

    ```

    my_env\Scripts\activate.bat

    ```

2. Install `Xvfb` to use `pyvirtualdisplay` for virtual display

    MacOS: Xvfb is not directly available on macOS. However, you can install XQuartz, which provides an X11 server that includes Xvfb functionality:

        

    - Download and install XQuartz from https://www.xquartz.org/.

3. Install `swig` for `box2d-py` and `gym[box2d]`

    MacOS:

    ```

    brew install swig

    ```

4. Install dependencies

    ```

    pip install -r requirements.txt

    ```

4. Start the program

    Run the `.ipynb` file on **Jupyter Notebook** or **Jupyer Lab**