Tutorials

These tutorials walk you through laser-measles step by step, from your first simulation to full vaccination modeling. All tutorials run in under one minute on a standard laptop — no cloud compute, GPU, or HPC cluster required.

Hardware requirements

Every tutorial is designed to run on a standard laptop or desktop with 4 GB of RAM and a modern Python installation (3.10+). If you are working in a setting with limited internet access, install laser-measles and its example dependencies while connected, then run tutorials offline.

Quick start

New to laser-measles? Start here. This single example walks through a complete spatial ABM from scenario construction to visualization.

• Quick start: Hello world

Beginning tutorials

These tutorials cover the fundamentals of building and running models. They introduce the three model types, the scenario format, and the component system.

• Basic model — build your first compartmental model
• Model structure — understand model internals and data flow
• ABM introduction — switch to agent-based modeling
• Scenarios — construct scenario DataFrames from geographic data

Feature tutorials

These tutorials highlight specific features of the library. Each one focuses on a single capability that you can combine with others.

• Random numbers — reproducible stochastic simulations
• Creating a component — write a custom process or tracker (how-to guide)
• ABM vital dynamics — births, deaths, and aging in agent-based models
• Spatial mixing — model disease spread between patches (concepts)
• Pydantic component parameters — validated configuration with Pydantic
• State arrays — efficient array operations on agent properties

End-to-end tutorials

These tutorials combine multiple features into realistic analysis workflows.

• Vaccination — model supplementary immunization activities and evaluate campaign impact

Calibration

Fitting model parameters to data. These tutorials assume you've already worked through a 1-node compartmental + Optuna example (parameter sweeps, loss design, basic Optuna usage).

• Spatial multi-modal calibration — progressive CMP→ABM cascade on a synthetic three-cluster chain. Expensive stages load from cached artifacts; cheap stages run live (~5 min for the CMP cold-start cell).
• Chain mixing visualizer (optional) — standalone companion to the spatial calibration tutorial. Visualizes the chain-mixer matrix, network, and population flow; validates a function-first reimplementation against the calibration tutorial's class via a numerical regression check. Self-contained, ~30 s to run.

Next steps

After completing the tutorials, explore these resources:

• Model types — explanation of the three model types and their trade-offs
• Choosing a model type — decision guide for selecting the right model
• Components — how the component system works
• Worked examples — copy-paste runnable scripts for all three models
• API reference — full details on all classes and functions

AI assistance

For those within IDM you can use our custom GPT Laser-GPT Jenner to get help with laser-core.