palaestrAI Architecture

palaestrAI has a distributed architecture at its core. The following documents help you to understand the internals of palaestrAI itself. This is useful if you want to contribute to palaestrAI itself.

Note

The following documents are not relevant for you if you are a user of palaestrAI (the core framework itself). This applies if you want to contribute an agent implementation, algorithm, environment, objectives, etc.

If you are interested in contributing new agent implementations, please have a look at the Brain-Muscle API.

Creating new environments is documented in the corresponding section on environments

• 1. Sequence diagram: A sequence diagram showing an experiment run
  • 1.1. Initialization
  • 1.2. Main Loop
  • 1.3. Shutdown
• 2. Run governor diagram: A diagram showing the different states of the :class:`~RunGovernor`
  • 2.1. Initialization and Startup
  • 2.2. Shutdown and clean up
  • 2.3. Error handling
• 3. Event State Machine
  • 3.1. Introduction and Purpose
  • 3.2. Class Documentation
    • EventStateMachine