Navigation

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

class WalkerList( list )	Source

WalkerList is a list of Walker.

It is the working ensemble of NestedSampler.

Attributes

• logZ : float (read-only)
       Natural log of evidence
• info : float (read-only)
       The information H. The compression factor ( the ratio of the prior space
       available to the model parameters over the posterior space ) is equal to the exp( H ).
• iteration : int
       Present iteration number.

Author Do Kester

WalkerList( problem=None, ensemble=0, allpars=None, fitIndex=None, walker=None, walkerlist=None )

Constructor.

To be valid it needs either problem/allpars/fitindex or walker or walkerlist

Parameters

• problem : Problem or None
       to construct a walker to be added.
• ensemble : int
       number of walkers
• allpars : array_like
       parameters of the problem
• fitIndex : array of int
       list of parameters to be fitted.
• walker : Walker or None
       walker to be added.
• walkerlist : Walkerlist or None
       walkerlist to be incorporated.

addWalkers( walker, ensemble )

setWalker( walker, index )

replace/append a Walker to this list

Parameters

• walker : Walker
       the list to take to copy from
• index : int
       the index at which to set

copy( src, des, wlist=None, start=0 )

Copy one item of the list onto another.

Parameters

• src : int
       the source item
• des : int
       the destination item
• wlist : WalkerList or None
       Copy from this WalkerList (None == self)
• start : int
       iteration where this walker was created

logPlus( x, y )

Return the log of sum.

firstIndex( lowL )

Return index of the first walker with walker.logL > lowL,
         None if list is empty
         len if no item applies

Parameters

• lowL : float
       low Likelihood

insertWalker( walker )

Insert walker to this list keeping it sorted in logL

Parameters

• walker : Walker
       the list to take to copy from

cropOnLow( lowL )

Return WalkerList with all LogL > lowL

Precondition: self is ordered on logL

Parameters

• lowL : float
       low Likelihood

getLogL( walker=None )

Return the logL of the/all walker

Parameters

• walker : None or Walker
       None return value for all walkers
       get the logL from

allPars( )

Return a 2d array of all parameters.

In case of dynamic models the number of parameters may vary. They are zero-padded. Use getNumberOfParametersEvolution to get the actual number.

Parameters

• kpar : int or tuple of ints
       the parameter to be selected. Default: all

getParameterEvolution( kpar=None )

Return the evolution of one or all parameters.

In case of dynamic models the number of parameters may vary. They are zero-padded. Use getNumberOfParametersEvolution to get the actual number.

Parameters

• kpar : int or tuple of ints
       the parameter to be selected. Default: all

getScaleEvolution( )

Return the evolution of the scale.

getLogLikelihoodEvolution( )

Return the evolution of the log( Likelihood ).

getLowLogL( )

Return the lowest value of logL in the walkerlist, plus its index.