Skip to content

Latest commit

 

History

History
87 lines (59 loc) · 5.35 KB

Installation-Guide.md

File metadata and controls

87 lines (59 loc) · 5.35 KB

Detailed Installation Guide

Table of Contents

Introduction

Welcome to the comprehensive installation guide for Animal-AI. This guide is designed to help you install and set up Animal-AI smoothly, even if you're not familiar with Python dependencies, GitHub repositories, or Unity.

For Windows Users: This guide is tailored for Windows, aiming to provide a straightforward installation process.

For Mac and Linux Users: Most instructions for Windows should apply to you as well. For MacOS, you might need to run chmod -R 777 AnimalAI.app in your terminal to adjust permissions. Linux users may need to make the .exe file executable with chmod +x env/AnimalAI.x86_64 and ensure files are moved to the parent folder after extraction.

Steps:

1. Installing Python

  • Download Python: Obtain Python 3.10.x from Python's official website. N.B. Python 3.10.x is the mininum version for AAI version 4.2.0 and above.
  • Run the Installer: Follow the installation prompts and ensure to add Python to your PATH (via the checkbox). For custom installations, keep the install pip box ticked.
  • Check Installation: Open Command Prompt and run python --version to verify the version. It should be Python 3.10.x.
  • Check Pip: Run pip --version. If pip is not installed, run python -m ensurepip. Pip is included by default in Python 3.4 and later.
  • Check PATH: Run echo %PATH% to verify Python is added to your PATH. Instructions for manual addition can be found here.

2. Cloning the Animal-AI Repository (Optional)

This step is only needed if you want to contribute to Animal-AI. For users who just want to use the environment, you can skip to step 3.

  • Prepare a Directory: Create a root folder for the AnimalAI project.
  • Clone the Repository: Choose one of the following options:
  • Verify: Ensure the root folder contains the animal-ai directory.

3. Setting Up a Virtual Environment (Optional)

  • Creating a Virtual Environment: This helps manage dependencies.
    • Python: Use python -m venv your_env_name and activate it from the Scripts directory with activate.
    • Conda: Use conda create --name your_env_name and activate with conda activate your_env_name.

For more on virtual environments, see the Python Documentation or Conda Documentation.

4. Installing Dependencies

  • Navigate to the Repository: Go to the animal-ai directory.
  • Install Dependencies:
    • Using pip: Run pip install animalai to install necessary dependencies from PyPI.
    • Using Conda: Install pip with conda install pip, then run pip install animalai.
    • Using a Virtual Environment: Activate your environment, then run pip install animalai.
  • Check: Run pip list to confirm animalai is installed. For a specific version, use pip install animalai==x.x.x (e.g. 4.0.0).

5. Downloading the Animal-AI Environment

  • Download: Get the appropriate version for your OS from the Releases section.
  • Extract: Unzip the downloaded file to your preferred location (Desktop recommended). Use tools like WinRAR or 7-Zip.
  • Verify: Ensure the folder contains the .exe file for Windows, .app file for MacOS, and .x86_64 file for Linux.

6. Starting Animal-AI

Launch the Animal-AI application from the directory where you extracted the files. Note that Animal-AI does not need to be installed on your system.

Refer to our Launching AAI guide for detailed instructions on starting and using Animal-AI.

General Notes

  • Folder Navigation: Use cd to change directories. For example, cd AAI moves to the "AAI" folder. Use cd.. to go up one directory level. For directory names with spaces, use quotes: cd "AAI Folder". List contents with dir or dir /b for a basic listing.

  • Dependencies: All necessary dependencies, including Unity's ml-agents package, are included in the Python package animalai. This is the only dependency needed. The package itself imports everything required.

Troubleshooting

If you encounter issues, please contact [email protected]. For common issues and solutions, visit our FAQ page.

Happy experimenting with Animal-AI!