Training SBI Models

In this section, we provide an overview of the various methods for training Simulation-Based Inference (SBI) models in Synference. Training an SBI model is a crucial step in the simulation-based inference workflow, as it allows us to learn the relationship between model parameters and synthetic observables from a library of pre-computed simulations or via online learning.

• Creating a Feature Array
• Basic SBI Model Training
• Feature and Parameter Arrays
• Training an SBI Model
• Loading a Trained Model
• Plotting model loss
• Plotting validation metrics
• Getting model metrics
• Posterior Samples
• Next Steps
• Complex SBI Model Training
• Model Validation
• Model Optimization
• Online Training
• Sampling Validation using MC