Class markov::Metropolis

Metropolis Algorithm.

Random sampling approximating the Boltzmann distribution using a Markov chain Monte Carlo approach.

Inheritance

markov::Walker

markov::Metropolis

Inherited Members

get_lowest_state

check_lowest

state

reset_evaluation_counter

make_sweep

swap_state

make_sweep

set_model

model

Walker

make_sweeps

make_sweep

attempt_transition

get_evaluation_counter

set_rng

save_lowest

rng

apply_transition

init

get_lowest_cost

make_step

cost

init

apply_scale_factor

compare

attempt_transition

apply_transition

verify_cost_difference

configure

render

~Component

Component

get_status

param

get_class_name

Constructors

Metropolis()

Create a Metropolis instance with uninitialized model and state.

Declaration

markov::Metropolis<Model>::Metropolis()

Methods

temperature()

Get the current temperature for sampling.

Declaration

double markov::Metropolis<Model>::temperature() const

beta()

Get the current inverse sampling temperature.

Declaration

double markov::Metropolis<Model>::beta() const

set_temperature()

Set the sampling temperature (also updates beta_).

Declaration

void markov::Metropolis<Model>::set_temperature(double temperature)

set_beta()

Declaration

void markov::Metropolis<Model>::set_beta(double beta)

accept()

Decide whether to accept a given cost increase.

Declaration

bool markov::Metropolis<Model>::accept(const typename Model::Cost_T&cost_diff) override

memory_estimate()

Estimate memory consumption using model parameters.

Declaration

static size_t markov::Metropolis<Model>::memory_estimate(const Model&model)