FitnessResult class abstract
This is an encoding of a phenotype's fitness. In simple cases, you can just use SingleObjectiveResult, which is just a single double, essentially.
But many scenarios require multi-dimensional fitness results. For example, a car can have various degrees of being comfortable, safe and fast. When comparing two cars, you don't want to just average the three traits into one. This is the idea behind Pareto fronts and Multi-objective optimization.
- Implemented types
- Implementers
Constructors
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- paretoRank ↔ int
-
Pareto rank of the fitness result. This is computed and assigned
in
GeneticAlgorithm._assignParetoRanks
.getter/setter pair - runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
compareTo(
FitnessResult other) → int -
Fitness results compare according to their paretoRank first and
then according to the result of evaluate.
override
-
dominates(
covariant FitnessResult other) → bool - A result dominates other results if it's better in every aspect.
-
evaluate(
) → double - Evaluates to a single numeric value. This goes against the multi-dimensionality of this class, but it's sometimes useful. For example, fitness sharing needs it, and it's easier for printing to the user.
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited