Package PyDSTool :: Package Toolbox :: Package optimizers :: Package criterion :: Module information_criteria :: Class ModifiedAICCriterion
[hide private]
[frames] | no frames]

Class ModifiedAICCriterion

source code

object --+
         |
        ModifiedAICCriterion

The Akaike information criterion with several trials authorized

Instance Methods [hide private]
 
__init__(self, ftol, iterations_max, correct_factor=1.0, trials=5)
Initializes the criterion with a max number of iterations and an error fraction for the monotony test...
source code
 
__call__(self, state, **kwargs)
Computes the stopping criterion
source code

Inherited from object: __delattr__, __getattribute__, __hash__, __new__, __reduce__, __reduce_ex__, __repr__, __setattr__, __str__

Properties [hide private]

Inherited from object: __class__

Method Details [hide private]

__init__(self, ftol, iterations_max, correct_factor=1.0, trials=5)
(Constructor)

source code 

Initializes the criterion with a max number of iterations and an error fraction for the monotony test
- ftol is the relative tolerance of the AIC criterion
- trials indicates how many time the criterion will return false when in fact the criterion was true
- correct_factor is the modifiying factor for the weight of the parameters size

Overrides: object.__init__