Pymc Bayesian Modeling

This gives you the full Bayesian modeling workflow with PyMC, from prior predictive checks through MCMC sampling to posterior diagnostics. It covers the practical stuff: hierarchical models with non-centered parameterization to avoid divergences, model comparison with LOO and WAIC, and all the diagnostics you need to check (R-hat, ESS, trace plots). The templates are solid for common patterns like logistic regression and time series. If you're doing uncertainty quantification or working with multilevel data where you need principled handling of missing observations, this walks you through the standard workflow without hand-waving the hard parts.