Which inference methods are supported?

The skill supports modern PyMC 5.x+ methods, including NUTS (No-U-Turn Sampler) for MCMC, and Automatic Differentiation Variational Inference (ADVI) for faster approximations.

What is the PyMC Bayesian Modeling skill?

This skill provides specialized guidance for Claude to help users build, sample, and diagnose Bayesian models using the PyMC Python library and the ArviZ diagnostic suite.

How does it handle model validation?

It guides users through the complete Bayesian workflow, including prior predictive checks, posterior predictive checks (PPC), and strict diagnostic evaluations like R-hat and Effective Sample Size (ESS).

Does this skill support hierarchical models?

Yes, it includes specific implementation patterns for multilevel modeling, emphasizing critical techniques like non-centered parameterization to ensure sampling convergence.

PyMC Bayesian Modeling

Name: PyMC Bayesian Modeling
Author: pathvanasoftware

bypathvanasoftware

数据科学与机器学习

Builds and validates sophisticated Bayesian probabilistic models using the PyMC library for advanced statistical inference.

关于

The PyMC Bayesian Modeling skill empowers developers and data scientists to implement robust probabilistic programming workflows directly within Claude Code. It provides domain-specific guidance for building hierarchical models, performing MCMC sampling with the NUTS sampler, and executing variational inference (ADVI). By integrating best practices for prior/posterior predictive checks and model diagnostics via ArviZ, this skill ensures statistical rigor in uncertainty quantification, missing data handling, and model comparison through LOO and WAIC metrics.

主要功能

Automated model diagnostics including R-hat, ESS, and divergence checks
Hierarchical and multilevel model construction with non-centered parameterization
Advanced MCMC sampling using the No-U-Turn Sampler (NUTS)
0 GitHub stars
Comprehensive posterior and prior predictive validation workflows
Model comparison and selection using LOO and WAIC information criteria

使用场景

Analyzing grouped or nested data structures through hierarchical modeling
Quantifying uncertainty in complex linear and logistic regression models
Forecasting and time series analysis using autoregressive Bayesian processes

关于

主要功能

Automated model diagnostics including R-hat, ESS, and divergence checks
Hierarchical and multilevel model construction with non-centered parameterization
Advanced MCMC sampling using the No-U-Turn Sampler (NUTS)
0 GitHub stars
Comprehensive posterior and prior predictive validation workflows
Model comparison and selection using LOO and WAIC information criteria

使用场景

Analyzing grouped or nested data structures through hierarchical modeling
Quantifying uncertainty in complex linear and logistic regression models
Forecasting and time series analysis using autoregressive Bayesian processes