Navigation

Global	Documentation
Home	Manual
Readme	Glossary
Notes	Design
Examples	Classes
	Trouble
	Style

class Walker( object )	Source

Walker is member of the cloud of points used in NestedSampler.

Attributes

• id : int
       identification number
• parent : int
       id of the parent (-1 for Adam/Eve)
• start : int
       iteration in which the walker is constructed
• step : int
       number of randomization steps since copy
• problem : Problem
       the problem being addressed
• logL : float
       log Likelihood = log Prob( data | params )
• logPrior : float
       log Prior for the model
• allpars : array_like
       list of parameters and hyperparameters
• fitIndex : array_like
       list of (super)parameters to be fitted.
• parameters : array_like (read only)
       parameters (of the model)
• hypars : array_like (read only)
       list of hyper parameters (of the error distribution)

Author Do Kester

Walker( wid, problem, allpars, fitIndex, logL=0, parent=-1, start=0, copy=None )

Constructor.

Either errdis or copy is obligatory.

Parameters

• wid : int
       id of the walker
• problem : Problem
       the problem being used. Parameters are copied from its model.
• allpars : array_like
       array of parameters and hyperparameters
• fitIndex : None or array_like
       indices of allpars to be fitted
       None is all
• logL : float
       log Likelihood
• parent : int
       id of the parent (-1 for Adam/Eve)
• start : int
       iteration in which the walker is constructed
• copy : Walker
       the walker to be copied

copy( )

Copy.

toSample( logW )

Return the contents of the Walker as a Sample.

check( nhyp=0, nuisance=0 )

Perform some sanity checks.