public abstract class AbstractParallelizableCostFunction extends AbstractSupervisedCostFunction<Vector,Vector> implements ParallelizableCostFunction
Constructor and Description |
---|
AbstractParallelizableCostFunction(java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>> costParameters)
Creates a new instance of AbstractParallelizableCostFunction
|
Modifier and Type | Method and Description |
---|---|
Vector |
computeParameterGradient(GradientDescendable function)
Differentiates function with respect to its parameters.
|
java.lang.Double |
evaluate(Evaluator<? super Vector,? extends Vector> evaluator)
Computes the cost of the given target.
|
clone, evaluatePerformance, getCostParameters, setCostParameters, summarize
evaluatePerformance
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
computeParameterGradientAmalgamate, computeParameterGradientPartial, evaluateAmalgamate, evaluatePartial
clone, getCostParameters, setCostParameters
summarize
evaluatePerformance
evaluatePerformance
public AbstractParallelizableCostFunction(java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>> costParameters)
costParameters
- Dataset to usepublic java.lang.Double evaluate(Evaluator<? super Vector,? extends Vector> evaluator)
CostFunction
evaluate
in interface Evaluator<Evaluator<? super Vector,? extends Vector>,java.lang.Double>
evaluate
in interface CostFunction<Evaluator<? super Vector,? extends Vector>,java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>>
evaluate
in class AbstractSupervisedCostFunction<Vector,Vector>
evaluator
- The object to evaluate.public Vector computeParameterGradient(GradientDescendable function)
DifferentiableCostFunction
computeParameterGradient
in interface DifferentiableCostFunction
function
- The object to differentiate.