Can I perform model comparison with this skill?

Yes, the skill includes templates for comparing multiple models using Leave-One-Out (LOO) cross-validation and WAIC, including Pareto-k reliability checks.

Which version of PyMC is supported by this skill?

This skill is specifically designed for PyMC version 5.x and later, utilizing the most recent API patterns and the ArviZ library for posterior analysis.

How does the skill handle MCMC sampling divergences?

The skill provides specific troubleshooting steps for divergences, including increasing target_accept values and implementing non-centered parameterization for hierarchical models.

Does it support Variational Inference (VI)?

Yes, it provides patterns for using Automatic Differentiation Variational Inference (ADVI) for quick exploration and initialization of MCMC chains.

PyMC Bayesian Modeling

Name: PyMC Bayesian Modeling
Author: BbgnsurfTech

byBbgnsurfTech

•

Data Science & ML

Builds, fits, and validates robust Bayesian statistical models using the PyMC probabilistic programming framework.

This skill provides a comprehensive workflow for performing Bayesian inference within Claude Code using the PyMC library. It guides users through the entire probabilistic programming lifecycle, including data preparation, weakly informative prior selection, hierarchical model construction, and high-performance MCMC sampling via NUTS. The skill emphasizes best practices like non-centered parameterization for multilevel models, rigorous diagnostic checks using R-hat and ESS, and model comparison through information criteria like LOO and WAIC, ensuring statistically sound results for complex data analysis tasks.

Key Features

01Support for hierarchical and non-centered parameterization

02Model comparison and selection using LOO/WAIC metrics

03Standardized Bayesian workflow from priors to posterior checks

04Integrated diagnostic scripts for convergence and sampling issues

053 GitHub stars

06Automated MCMC (NUTS) and Variational Inference (ADVI) setups

Use Cases

01Analyzing time-series data with autoregressive Bayesian processes

02Modeling multi-level grouped data with hierarchical structures

03Estimating uncertainty in complex linear and logistic regressions

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bbgnsurftech/claude-skills-collection pymc

For use in Claude.ai and ChatGPT

Download Skill