Schola Documentation

Schola Documentation

The Schola project is an effort to build a toolkit/plugin for controlling Objects in Unreal with Reinforcement Learning. It provides tools to help the user create Environments, define Agents, connect to python based RL Frameworks (e.g. Gym, RLLib or Stable Baselines 3), and power NPCs with RL during games.

• Guides

  • Getting Started with Schola
    • Install Prerequisites
    • Install Schola
    • Building 3rd-Party Dependencies for Schola (Optional)
  • Running Schola
    • Launch An Unreal Environment From Python
    • Connect To a Running Unreal Environment
  • Setting Up Inference
    • Convert a Checkpoint to Onnx
    • Load an Onnx Model into Unreal Engine
    • Setting up Your Unreal Engine Level
  • Running with a Standalone Environment from CLI
    • Building Your Environment
    • Running From CLI
  • Tips and Tricks
    • Training In The Editor Runs Slowly
    • Getting More CPU/GPU Utilization
    • Training Runs at a High FPS but My Agent Doesn’t Train
    • Training is Complete but Training Script Hangs
    • Understanding Algorithm Parameters and Options
  • Migrating from Schola V1.3 to V2
    • Overview of Changes
    • Python API Changes
    • Unreal Engine API Changes
    • Imitation Learning
    • Migration Checklist
    • Common Migration Issues
    • Additional Resources

• Examples

  • Training NPCs to Play a MultiAgent Game of Tag
    • The Structure of the Environment in Unreal Engine
    • Initial Setup
    • Creating the Custom Direction and Distance Observer
    • Creating the Agents
    • Creating the Environment Definition
    • Creating the Map
    • Registering the Agents
    • Starting Training
  • Training an X-Arm 5 Robotic Arm with AMD Schola and Unreal Engine
    • Table of Contents
    • Getting Started
    • Training Tasks
    • Prerequisites

• API Documentation

  • Python
    • Core
    • Extensions
    • Scripts
    • Generated gRPC Code
  • Unreal