Package criterion

source code

Module containing every criteria for converge test

Functions :

crietrion() creates a composite criterion

Criteria :

IterationCriterion
- stops when the iteration limit is reached
MonotonyCriterion
- stops when the cost rises again
RelativeValueCriterion
- stops when the relative value error is below a certain level
AbsoluteValueCriterion
- stops when the absolute value error is below a certain level
RelativeParametersCriterion
- stops when the relative parameters error is below a certain level
AbsoluteParametersCriterion
- stops when the absolute parameters error is below a certain level
GradientCriterion
- stops when the gradient is below a certain level

Composite criteria :

OrComposition
- returns True if one of the criteria returns True
AndComposition
- returns True if all the criteria return True

Information criteria :

AICCriterion
- stops when the cost function increases again
- is dependent of the number of parameters if it changes
ModifiedAICCriterion
- Identical to AICCriterion except that a number of increases are tolerated

Submodules

[hide private]

PyDSTool.Toolbox.optimizers.criterion.composite_criteria: Composite criteria allow to use several criteria together, with and/or composition
PyDSTool.Toolbox.optimizers.criterion.criteria: A list of standard convergence criteria based on the number of iterations, the last values taken by the cost function and the associated points
PyDSTool.Toolbox.optimizers.criterion.facilities: Proposes a way to create a composite criterion
PyDSTool.Toolbox.optimizers.criterion.information_criteria

Classes

[hide private]

RelativeValueCriterion
The relative criterion stops the optimization when the relative error of the value is below a certain level (ftol)

GradientCriterion
The gradient criterion stops the optimization when the gradient at the current point is less that a given tolerance

AbsoluteValueCriterion
The absolute criterion stops the optimization when the absolute error of the value is below a certain level (ftol)

AbsoluteParametersCriterion
The absolute criterion stops the optimization when the relative error of the parameters is below a certain level (xtol)

RelativeParametersCriterion
The relative criterion stops the optimization when the relative error of the parameters is below a certain level (xtol)

IterationCriterion
A simple criterion that stops when the iteration limit is reached

MonotonyCriterion
A simple criterion that stops when the values of the function starts to rise again

AndComposition
Compose several criteria with an and rule

OrComposition
Compose several criteria with an or rule

AICCriterion
The Akaike information criterion

ModifiedAICCriterion
The Akaike information criterion with several trials authorized

Functions

[hide private]

criterion(**kwargs)
Creates a composite criterion based on the formal parameters :

source code

Variables

[hide private]

criterion__all__ = ['IterationCriterion', 'MonotonyCriterion',...

Function Details

[hide private]

criterion(**kwargs)

source code

Creates a composite criterion based on the formal parameters :

iterations_max indicates the maximum number of iteration
ftol is the maximum relative change of the value function
xtol is the maximum relative change of the parameters
gtol is the maximum gradient

Variables Details

[hide private]

criterionall

Value:

['IterationCriterion',
 'MonotonyCriterion',
 'RelativeValueCriterion',
 'AbsoluteValueCriterion',
 'RelativeParametersCriterion',
 'AbsoluteParametersCriterion',
 'GradientCriterion',
 'OrComposition',
...