Module bayesian_inference 

Bayesian inference: priors, likelihoods, posterior updates, MCMC, Bayesian linear regression, Gaussian processes, and model selection.

All distributions work with real-valued parameters; conjugate update formulas are provided where analytic posteriors exist.

Structs

BayesianLinearRegression

Bayesian linear regression with conjugate Gaussian prior on weights.

BayesianUpdate

Conjugate Bayesian updates for standard distribution families.

GaussianProcess

Gaussian Process regression with a scalar-input, scalar-output model.

MarkovChainMonteCarlo

Markov Chain Monte Carlo samplers.

ModelSelection

Model selection criteria: AIC, BIC, Bayes factor, and cross-validation.

Enums

Kernel

Kernel function types for Gaussian processes.

Likelihood

Supported likelihood functions.

Prior

Supported prior distribution families.

Functions

log_beta

Returns ln B(a, b) = ln Γ(a) + ln Γ(b) - ln Γ(a+b).

log_gamma

Returns ln Γ(x) using the Lanczos approximation.