Module: tfp.substrates.numpy.mcmc

TensorFlow Probability MCMC python package.

Classes

class CheckpointableStatesAndTrace: States and auxiliary trace of an MCMC chain.

class DualAveragingStepSizeAdaptation: Adapts the inner kernel's step_size based on log_accept_prob.

class MetropolisHastings: Runs one step of the Metropolis-Hastings algorithm.

class RandomWalkMetropolis: Runs one step of the RWM algorithm with symmetric proposal.

class ReplicaExchangeMC: Runs one step of the Replica Exchange Monte Carlo.

class SimpleStepSizeAdaptation: Adapts the inner kernel's step_size based on log_accept_prob.

class SliceSampler: Runs one step of the slice sampler using a hit and run approach.

class StatesAndTrace: States and auxiliary trace of an MCMC chain.

class TransformedTransitionKernel: TransformedTransitionKernel applies a bijector to the MCMC's state space.

class TransitionKernel: Base class for all MCMC TransitionKernels.

class UncalibratedRandomWalk: Generate proposal for the Random Walk Metropolis algorithm.

Functions

default_swap_proposal_fn(...): Make the default swap proposal func, with P[swap], for replica swap MC.

effective_sample_size(...): Estimate a lower bound on effective sample size for each independent chain.

even_odd_swap_proposal_fn(...): Make a deterministic swap proposal function, alternating even/odd swaps.

potential_scale_reduction(...): Gelman and Rubin (1992)'s potential scale reduction for chain convergence.

random_walk_normal_fn(...): Returns a callable that adds a random normal perturbation to the input.

random_walk_uniform_fn(...): Returns a callable that adds a random uniform perturbation to the input.

sample_annealed_importance_chain(...): Runs annealed importance sampling (AIS) to estimate normalizing constants.

sample_chain(...): Implements Markov chain Monte Carlo via repeated TransitionKernel steps.

sample_halton_sequence(...): Returns a sample from the dim dimensional Halton sequence.