MCScheduling 1.0
Set of Algorithms for Solving Mixed-Criticality Scheduling
|
Sigma scaling may be used to force a constant selection pressure over many generations. More...
Public Member Functions | |
void | scale (CPopulation population) |
Scales the fitness scores of the individuals within the specified population. |
Sigma scaling may be used to force a constant selection pressure over many generations.
It works like this: First, the average fitness score - averageFitness - is calculated, and then the standard deviation of the fitness scores - sigma - is calculated. Eventually a new fitness score for each individual in the specified population is assigned according to this equation: newFitnessScore = (oldFitnessScore - averageFitness) / (2 * sigma) If the sigma is 0, then all the fitness scores are set to 1.0.
void MCScheduling.GeneticAlgorithm.CSigmaScaling.scale | ( | CPopulation | population | ) |
Scales the fitness scores of the individuals within the specified population.
For more information look at the description of the CSigmaScaling class.
population | The population of individuals to be scaled. |
Implements MCScheduling.GeneticAlgorithm.IFitnessScaler.