Module: oryx.experimental.mcmc.kernels
bookmark_border Stay organized with collections Save and categorize content based on your preferences.

Contains probabilistic program kernels for MCMC.

Functions

hmc(...): Makes a Hamiltonian Monte Carlo step function.

mala(...): Makes a Metropolis-adjusted Langevin algorithm (MALA) step function.

metropolis(...): Returns a program that takes a Metropolis step with an inner kernel.

metropolis_hastings(...): Returns a program that takes a Metropolis-Hastings step.

random_walk(...): Returns a probabilistic program that takes a Gaussian step with a provided variance.

sample_chain(...): Runs several steps of MCMC.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.

Last updated 2024-05-23 UTC.