Home

Bringing Machine Learning to Unreal Engine

InteractML brings machine learning visual scripting to Unreal, empowering creators to develop novel gameplay mechanics, visual experiences, and control schemes without writing code. You can create machine learning models by joining Blueprint nodes together, feeding them from any available parameters, and then using the outputs to drive any in-engine systems or effects you like.

Contents

1. Introduction - About the project, the plugin, and some typical use cases
2. Setup - Getting set up, plugin installation, configuration, first steps
3. Machine Learning - Brief info on the ML systems available with Interact ML and what they do
4. Blueprint Node Types - Overview of the Blueprint nodes available and common features
5. Asset Types - The model, training set, and label asset types and what they are for
6. Recording Examples - How to record example data
7. Training Models - How to train a model
8. Running Models - How to run a model
9. Model Types - Model specific behaviour and use
10. Training Set - Training set specifics, including editing window
11. Labels - Label and Label Table asset specifics and use
12. Data - How and where the example and model data is stored and what you can do with it
13. Demo Project - Information on all the demos available, what they show, and how they are set up
14. Blueprints - Details of the demo blueprints, helpers, interaction, state, utility
15. Technical - Some technical/implementation details, diagrams, and designs