Trait net_ensembles::sampling::entropic_sampling::Entropic[−][src]

pub trait Entropic {
    fn step_counter(&self) -> usize;

    fn total_steps_accepted(&self) -> usize;

    fn total_steps_rejected(&self) -> usize;

    fn step_goal(&self) -> usize;

    fn log_density(&self) -> Vec<f64, Global>;

    fn write_log<W>(&self, writer: W) -> Result<(), Error>
    where
        W: Write;

    fn steps_total(&self) -> usize { ... }

    fn fraction_accepted_total(&self) -> f64 { ... }

    fn fraction_rejected_total(&self) -> f64 { ... }

    fn is_finished(&self) -> bool { ... }

    fn log_density_base10(&self) -> Vec<f64, Global> { ... }

    fn log_density_base(&self, base: f64) -> Vec<f64, Global> { ... }
}
Expand description

Traits for quantities that all Entropic simulations have

Required methods

Counter

how many Entropic steps were performed until now?

How many steps were accepted until now?

  • this includes steps, that were perfored to find the estimate, i.e., during wang landau

How many steps were rejected until now?

  • this includes steps, that were perfored to find the estimate, i.e., during wang landau

What is the goal to reach?

Current (non normalized) estimate of ln(P(E))

  • i.e., of the natural logarithm of the probability density function for the requested interval
  • this is what we are doing the simulations for

Writes Information about the simulation to a file. E.g. How many steps were performed.

Provided methods

Counter

  • how many markov steps were performed until now?
  • this includes steps, that were perfored to find the estimate, i.e., during wang landau

Calculate, which fraction of steps were accepted

  • this includes steps, that were perfored to find the estimate, i.e., during wang landau
  • if no steps were performed, it returns f64::NAN

Calculate, which fraction of steps were rejected

  • this includes steps, that were perfored to find the estimate, i.e., during wang landau
  • if no steps were performed, it returns f64::NAN

Checks wang landau threshold

  • log_f <= log_f_threshold

Current (non normalized) estimate of log10(P(E))

  • i.e., of logarithm with base 10 of the probability density function for the requested interval
  • this is what we are doing the simulations for

Current (non normalized) estimate of log_base(P(E))

  • i.e., of logarithm with arbitrary base of the probability density function for the requested interval
  • this is what we are doing the simulations for

Implementors