Tutorials for using the sbi toolbox

Before running the notebooks, follow our instructions to install sbi. Alternatively, you can also open a codespace on GitHub and work through the tutorials in the browser. The numbers of the notebooks are not informative of the order, please follow this structure depending on which group you identify with.

Once you have familiarised yourself with the methods and identified how to apply SBI to your use case, ensure you work through the Diagnostics tutorials linked below, to identify failure cases and assess the quality of your inference.

Introduction

• Getting started
• Amortized inference
• More flexibility for training and sampling
• Implemented algorithms

Advanced

• Multi-round inference
• Sampling algorithms in sbi
• Custom density estimators
• Embedding nets for observations
• SBI with trial-based data
• Handling invalid simulations
• Crafting summary statistics
• Importance sampling posteriors

Diagnostics

• Posterior predictive checks
• Simulation-based calibration
• Local-C2ST coverage checks
• Density plots and MCMC diagnostics with ArviZ

Analysis

• Conditional distributions
• Posterior sensitivity analysis
• Plotting functionality

Examples

• Hodgkin-Huxley model
• Decision-making model