bayesflow-adapter

What problem does it solve?

This Skill streamlines the creation and debugging of data preprocessing pipelines for BayesFlow, ensuring simulator output is correctly formatted for inference and summary networks.

Core Features & Use Cases

• Schema Mapping: Organizes simulator output into inference_variables, summary_variables, and inference_conditions.
• Transformations: Applies custom element-wise or cross-key transformations to data.
• Standardization: Implements robust standardization using prior moments to prevent distribution shift.
• Use Case: You need to prepare simulation data for a Bayesian neural network. This Skill helps you define the exact preprocessing steps, including feature engineering and standardization, to feed into your model.

Quick Start

Use the bayesflow-adapter skill to create a pipeline that standardizes the 'param' key using prior moments.