Dynamic = class Dynamic(builtins.object) |
|
Dynamic(dynamic=True)
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 |
|
Constructor:
- Dynamic( dynamic=True )
- Constructor for Dynamic
Parameters
----------
dynamic: bool
True if the Model is to be considered dynamic.
Methods defined here:
- 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
- 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 .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)
- 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
- isDynamic()
- 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
- 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
- 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
- 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
|
|