`altar.bayesian.Brent`

Module Contents

Classes

class altar.bayesian.Brent.Brent(name, locator, **kwds)

Bases: altar.component

A δβ solver based on the GSL implementation of the Brent algorithm

tolerance

doc = the fractional tolerance for achieving convergence

maxiter

doc = the maximum number of iterations while looking for a δβ

cov

initialize(self, application, scheduler): Initialize me and my parts given an {application} context and a {scheduler}

solve(self, llk, weight): Compute the next temperature in the cooling schedule :param llk: data log-likelihood :param weight: the normalized weight :return: β, cov