alterlab-pymc

What problem does it solve?

Bayesian modeling workflows for researchers require structured guidelines, templates, and diagnostics to build, fit, and validate probabilistic models with PyMC.

Core Features & Use Cases

• End-to-end Bayesian workflow: from data preparation and model construction to sampling, diagnostics, and posterior predictive checks.
• Templates & references: ready-to-use patterns for linear, hierarchical, and time-series models, plus reference materials.
• Educational examples: practical use cases and tutorials for posterior analysis, model comparison, and predictions.
• Use Case: a researcher compares multiple models (linear vs hierarchical) on a dataset and selects the best via LOO/WAIC.

Quick Start

Load your dataset and run a PyMC Bayesian model to obtain posterior estimates.