Unity ML-Agents Toolkit Documentation

Installation & Set-up

• Installation
  • Using Virtual Environment

Getting Started

• Getting Started Guide
• ML-Agents Toolkit Overview
  • Background: Unity
  • Background: Machine Learning
  • Background: PyTorch
• Example Environments

Creating Learning Environments

• Making a New Learning Environment
• Designing a Learning Environment
  • Designing Agents
• Using an Executable Environment
• ML-Agents Package Settings

Training & Inference

• Training ML-Agents
  • Training Configuration File
  • Using TensorBoard to Observe Training
  • Profiling Trainers
• Unity Inference Engine

Extending ML-Agents

• Creating Custom Side Channels
• Creating Custom Samplers for Environment Parameter Randomization

Python Tutorial with Google Colab

• Using a UnityEnvironment
• Q-Learning with a UnityEnvironment
• Using Side Channels on a UnityEnvironment

Help

• Migrating from earlier versions of ML-Agents
• Frequently Asked Questions
• ML-Agents Glossary
• Limitations

API Docs

• API Reference
• Python API Documentation
• How to use the Python API
• How to use the Unity Environment Registry
• Wrapping Learning Environment as a Gym (+Baselines/Dopamine Integration)

Translations

To make the Unity ML-Agents Toolkit accessible to the global research and Unity developer communities, we're attempting to create and maintain translations of our documentation. We've started with translating a subset of the documentation to one language (Chinese), but we hope to continue translating more pages and to other languages. Consequently, we welcome any enhancements and improvements from the community.

• Chinese
• Korean

Deprecated Docs

We no longer use them ourselves and so they may not be up-to-date. We've decided to keep them up just in case they are helpful to you.

• Windows Anaconda Installation
• Using Docker
• Training on the Cloud with Amazon Web Services
• Training on the Cloud with Microsoft Azure
• Using the Video Recorder