Navigation

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

class Dynamic( object )	Source

Class adjoint to Model which implements some dynamic behaviour.

Attributes

• ncomp : int
       the number of components in the dynamic model
• deltaNpar : int
       the number of parameters in each component
• minComp : int
       minimum number of repetitions
• maxComp : None or int
       maximum number of repetitions
• growPrior : None or Prior
       governing the birth and death.
       ExponentialPrior (scale=2) if maxOrder is None else UniformPrior

Dynamic( dynamic=True )

Constructor for Dynamic

Parameters

• dynamic : bool
       True if the Model is to be considered dynamic.

isDynamic( )

setGrowPrior( growPrior=None, min=1, max=None, name="Comp" )

Set the growth prior.

Parameters

• growPrior : None or Prior
       governing the birth and death.
       ExponentialPrior (scale=2) if maxOrder is None else UniformPrior
• min : int
       lower limit on growthprior
• max : None or int
       upper limit on growthprior
• name : str
       name of the component

setDynamicAttribute( name, value )

Set attribute, if it belongs to a Dynamic Models.

Parameters

• name : str
       name of the attribute
• value : anything
       value of the attribute

Return

• bool : True if name was a Dynamic name
          False if not

grow( offset=0, rng=None, **kwargs )

Increase the degree by one upto maxComp ( if present ).

Parameters

• offset : int
       index where the params of the Dynamic model start
• rng : random number generator
       to generate a new parameter.

Return

• bool : succes

shrink( offset=0, rng=None, **kwargs )

Decrease the degree by one downto minComp ( default 1 ).

Parameters

• offset : int
       index where the params of the Dynamic model start
• rng : random number generator
       Not used in this implementation

Return

• bool : succes

alterParameterNames( dnp )

Renumber the parameter names.

Parameters

• dnp : int
       change in the number of parameters

alterParameterSize( dnp, offset, location=None, value=0 )

Change the number of parameters and self.parameters.

Parameters

• dnp : int
       change in the number of parameters in the DynamicModel
• offset : int
       starting index of the DynamicModel
• location : int
       index in param[offset:] at which to insert/delete the new parameters

alterParameters( param, location, dnp, offset, value=None )

change the parameters to comply with the changed model.

param: [p0 p1 p2 p3 p4 p5 p6 p7 p8 p9] # previous set offset: 2 # parameters of models in preceeding chain location: 1 # location where to add/delete parameter value: [v0 ...] # values to be given to added parameters

dnp: +1 ==> newpar: [p0 p1 p2 v0 p3 p4 p5 p6 p7 p8 p9]

dnp: +2 ==> newpar: [p0 p1 p2 v0 v1 p3 p4 p5 p6 p7 p8 p9]

dnp: -1 ==> newpar: [p0 p1 p3 p4 p5 p6 p7 p8 p9]

dnp: -2 ==> ERROR: not enough space in param before location

Parameters

• param : array_like
       parameters of the parent model (chain)
• location : int
       index in param[offset:] at which to insert/delete the new parameters
• dnp : int
       number of parameters to insert (dnp>0) or delete (dnp<0)
• offset : int
       start index of the parameters of the dynamic model in param
• value : float or array_like
       to be given to the inserted parameters (only when dnp>0)

alterFitindex( findex, location, dnp, offset )

change the fit index to comply with the changed model.

Parameters

• findex : array_like
       fit index of the parent model (chain)
• location : int
       index in param[offset:] at which to insert/delete the new parameters
• dnp : int
       number of parameters to insert (dnp>0) or delete (dnp<0)
• offset : int
       start index of the parameters of the dynamic model in param

shuffle( param, offset, np, rng )

Shuffle the parameters of the components (if they are equivalent) Default implementation: does nothing.

Parameters

• param : array-like
       list of all parameters
• offset : int
       index where the dynamic model starts
• np : int
       length of the parameters of the dynamic model
• rng : RNG
       random number generator