/Nyan_Nyan_Neko

Cat-Training-Simulator is a networked game where users create and train virtual cats with ML-Agents. Users can interact, teach, and customize their cats, and share them online. The game is made with Unity and Python, and works on Windows, Mac, and Linux.

MIT LicenseMIT

Nyan Nyan Neko

Nyan Nyan Neko Gēmu (meow meow cat game) is a network-based simulation game that allows users to create and train their own virtual cats in a realistic 3D environment.

Features

  • Interact with your cats, teach them new skills, and customize their appearance and personality.
  • Connect with other players and share your cats online.
  • Use the ML-Agents framework to enable machine learning and artificial intelligence for the cat agents.

Requirements

  • Unity 2022.3 or later
  • Python 3.10.12 or higher
  • ML-Agents Toolkit
  • Optional: ML-Agents Extensions and Plugin Examples

Installation

  • Clone the ML-Agents Toolkit repository from GitHub:
git clone --branch main https://github.com/Unity-Technologies/ml-agents.git
  • Install the com.unity.ml-agents Unity package using the Package Manager in Unity.
  • Install the mlagents Python package using pip:
pip install mlagents
  • Optional: Install the com.unity.ml-agents.extensions Unity package and the mlagents-plugin-examples Python package for additional features.

Usage

  • Create a new Unity project and import the ML-Agents packages.
  • Create your own cat agent prefab or use one of the provided examples.
  • Create your own training environment or use one of the provided examples.
  • Configure your training settings and run the trainer using the mlagents-learn command.
  • Enjoy your cat-training-simulator!

License

This project is licensed under the MIT License - see the LICENSE file for details.