Skip to content

Maze documentation

Maze documentation

Installation
A First Example
Maze - Step by Step
API Documentation

Workflow

Training
Rollouts
Deployment
Collecting and Visualizing Rollouts
Imitation Learning and Fine-Tuning
Experiment Configuration

Policy and Value Networks

Introducing the Perception Module
Action Spaces and Distributions
Working with Template Models
Working with Custom Models

Training

Maze Trainers

Concepts and Structure

Policies, Critics and Agents
Maze Environment Hierarchy
Maze Event System
Configuration with Hydra
Environment Rendering
Structured Environments

Environment Customization

Customizing Core and Maze Envs
Customizing / Shaping Rewards
Environment Wrappers
Observation Pre-Processing
Observation Normalization

Best Practices and Tutorials

Tricks of the Trade
Cheat Sheet
Integrating an Existing Gym Environment
Structured Environments and Action Masking
Combining Maze with other RL Frameworks
Plain Python Training Example (low-level)

Logging and Monitoring

Tensorboard and Command Line Logging
Event and KPI Logging
Action Distribution Visualization
Observation Logging

Scaling the Training Process

Runner Concept

MazeActionType¶

maze.core.env.maze_action.MazeActionType¶: alias of Any

RewardAggregatorInterface

Copyright © 2025, EnliteAI GmbH

Made with Sphinx and @pradyunsg's Furo

On this page

MazeActionType
- MazeActionType