Why should I use prior moments for standardization?

Analytical prior moments are deterministic and constant, preventing distribution shifts that occur when using empirical batch statistics, especially in models with variable-sized data.

What is the correct order for an adapter pipeline?

The recommended order is: 1. Type conversion, 2. Cross-key transforms, 3. Per-key transforms, 4. Standardization, 5. Concatenation, and 6. Dropping unused keys.

What is a BayesFlow adapter?

A BayesFlow adapter is a preprocessing pipeline that transforms raw simulator output into the specific schema (inference_variables, summary_variables, and inference_conditions) required by summary and inference networks.

How does this skill handle custom transformations?

The skill provides specific patterns for subclassing ElementwiseTransform for single keys and Transform for cross-key reparameterizations, ensuring they integrate correctly with the pipeline.

BayesFlow Adapter Assistant

Name: BayesFlow Adapter Assistant
Author: matthiaskloft

bymatthiaskloft

0•

데이터 과학 및 ML

Streamlines the creation and debugging of data preprocessing pipelines for BayesFlow simulation-based inference models.

The BayesFlow Adapter skill provides specialized guidance for building and optimizing the data pipelines that bridge simulator outputs with neural network architectures. It ensures correct implementation of the three-slot schema—inference variables, summary variables, and inference conditions—while enforcing critical best practices such as transforming data before standardization and using analytical prior moments instead of empirical statistics. This skill is essential for researchers and machine learning engineers working on simulation-based inference who need to avoid common pitfalls like train/validation distribution shifts and dimension mismatches.

주요 기능

010 GitHub stars

02Implements analytical prior moments for robust data standardization

03Provides templates for custom elementwise and cross-key transformations

04Maps simulator outputs into standardized inference, summary, and condition slots

05Enforces optimal pipeline ordering (transform before standardize)

06Supports declarative AdapterSpec patterns for HPO-compatible workflows

사용 사례

01Reparameterizing simulator outputs for improved neural network convergence

02Building robust preprocessing pipelines for simulation-based inference tasks

03Debugging dimension mismatches and distribution shifts in ML models

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add matthiaskloft/claude-skills bayesflow-adapter

For use in Claude.ai and ChatGPT

주요 기능

010 GitHub stars

02Implements analytical prior moments for robust data standardization

03Provides templates for custom elementwise and cross-key transformations

04Maps simulator outputs into standardized inference, summary, and condition slots

05Enforces optimal pipeline ordering (transform before standardize)

06Supports declarative AdapterSpec patterns for HPO-compatible workflows

사용 사례

01Reparameterizing simulator outputs for improved neural network convergence

02Building robust preprocessing pipelines for simulation-based inference tasks

03Debugging dimension mismatches and distribution shifts in ML models