StepSizeRules package

Module StepSizeRules

This module contains all the step size rules type that are provided by pyHarm. The module is organized around an abstract class ABCStepSizeRule and a FactoryStepSize that is in charge of creating the objects. All step size rule object must comply with the ABCStepSizeRule abstract class. The section below presents the different classes that are available in this module.

ABCStepSizeRule

The ABCStepSizeRule class is an abstract class defining the essential components of any step size rule. One abstract method is defined :

Methods	Use
ProjectInBounds	Projects the step size over the bounds
getStepSize	Abstract method : Returns the step size based on the previous step size and the list of SystemSolution provided

Examples of creating an ABCStepSizeRule and adding it into an input dictionary:

To be created, an ABCStepSizeRule subclass needs its abstract method to be defined :

class FakeStep(ABCStepSizeRule): # inherits from abstract class
    factory_keyword="fakestep" # mandatory to define
    def getStepSize(self, ds:float, sollist:list[SystemSolution], **kwargs) -> float:
        # Codes when to return True when condition is reached
        return ds

INP = {
    "analysis":{
        "FRF":{
            "study":"frf",
            ...,
            "stepsizer":"fakestep", # calls the criterion using factory_keyword.
            ...,
        },
        ...,
    },
    "plugin":[FakeStep], # adds to the factory using plugin
    ...,
}

FactoryStepSize

This file contains the dictionary of all the step size rules that are available as well as the function generateStepSizeRule that creates the ABCStepSizeRule object based on the type of corrector desired.

StepSizeConstant constant

The step size remains constant by all means. Be careful to use a stop criterion that is compliant with this step size rule.

StepSizeAcceptance acceptance

This rule returns a step size that is divided by two if the previous solution has not been accepted. Or multiply its value per 2 if a number of consecutive successes by the solver has been made (default is 5). Step size is projected over the defined bounds before being returned.

API links

pyHarm.StepSizeRules