io.github.ericmedvet.jgea.problem.classification.ClassificationFitness Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of jgea.problem Show documentation
Problem (benchmarks and templates) for jgea.
The newest version!
/*-
 * ========================LICENSE_START=================================
 * jgea-problem
 * %%
 * Copyright (C) 2018 - 2024 Eric Medvet
 * %%
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =========================LICENSE_END==================================
 */
package io.github.ericmedvet.jgea.problem.classification;

import io.github.ericmedvet.jgea.core.fitness.ListCaseBasedFitness;
import io.github.ericmedvet.jgea.core.util.LinkedHashMultiset;
import io.github.ericmedvet.jgea.core.util.Multiset;
import io.github.ericmedvet.jnb.datastructure.Pair;
import java.util.ArrayList;
import java.util.List;
import java.util.function.Function;

public class ClassificationFitness>
    extends ListCaseBasedFitness, O, L, List> {

  private final List> data;

  public ClassificationFitness(List> data, Metric errorMetric) {
    super(
        data.stream().map(Pair::first).toList(),
        Classifier::classify,
        getAggregator(data.stream().map(Pair::second).toList(), errorMetric));
    this.data = data;
  }

  public enum Metric {
    CLASS_ERROR_RATE,
    ERROR_RATE,
    BALANCED_ERROR_RATE
  }

  private record ClassErrorRate>(List actualLabels)
      implements Function, List>> {

    @SuppressWarnings("unchecked")
    @Override
    public List> apply(List predictedLabels) {
      E protoLabel = actualLabels.getFirst();
      E[] allLabels = (E[]) protoLabel.getClass().getEnumConstants();
      Multiset counts = new LinkedHashMultiset<>();
      Multiset errors = new LinkedHashMultiset<>();
      for (int i = 0; i < actualLabels.size(); i++) {
        counts.add(actualLabels.get(i));
        if (!actualLabels.get(i).equals(predictedLabels.get(i))) {
          errors.add(actualLabels.get(i));
        }
      }
      List> pairs = new ArrayList<>(allLabels.length);
      for (E currentLabel : allLabels) {
        pairs.add(new Pair<>(errors.count(currentLabel), counts.count(currentLabel)));
      }
      return pairs;
    }
  }

  private static > Function, List> getAggregator(
      List actualLabels, Metric metric) {
    final ClassErrorRate classErrorRate = new ClassErrorRate<>(actualLabels);
    if (metric.equals(Metric.CLASS_ERROR_RATE)) {
      return (List predictedLabels) -> {
        List> pairs = classErrorRate.apply(predictedLabels);
        return pairs.stream()
            .map(p -> ((double) p.first() / (double) p.second()))
            .toList();
      };
    }
    if (metric.equals(Metric.ERROR_RATE)) {
      return (List predictedLabels) -> {
        List> pairs = classErrorRate.apply(predictedLabels);
        int errors = pairs.stream()
            .map(Pair::first)
            .mapToInt(Integer::intValue)
            .sum();
        int count = pairs.stream()
            .map(Pair::second)
            .mapToInt(Integer::intValue)
            .sum();
        return List.of((double) errors / (double) count);
      };
    }
    if (metric.equals(Metric.BALANCED_ERROR_RATE)) {
      return (List predictedLabels) -> {
        List> pairs = classErrorRate.apply(predictedLabels);
        return List.of(pairs.stream()
            .map(p -> ((double) p.first() / (double) p.second()))
            .mapToDouble(Double::doubleValue)
            .average()
            .orElse(Double.NaN));
      };
    }
    return null;
  }

  public ClassificationFitness changeMetric(Metric metric) {
    return new ClassificationFitness<>(data, metric);
  }
}