場景：從樣本集中采樣80%用于訓練，20%用于驗證。

參考代碼如下：

package com.gddx;import java.io.File; import java.util.Map;import libsvm.LibSVM; import net.sf.javaml.classification.Classifier; import net.sf.javaml.classification.evaluation.EvaluateDataset; import net.sf.javaml.classification.evaluation.PerformanceMeasure; import net.sf.javaml.core.Dataset; import net.sf.javaml.sampling.Sampling; import net.sf.javaml.tools.data.FileHandler; import be.abeel.util.Pair;/*** Sample program illustrating how to use sampling.* * @author Thomas Abeel* */ public class TutorialSampling {public static void main(String[] args) throws Exception {Dataset data = FileHandler.loadDataset(new File("D:\\tmp\\javaml-0.1.7-src\\UCI-small\\iris\\iris.data"), 4, ",");Sampling s = Sampling.SubSampling;Pair<Dataset, Dataset> datass = s.sample(data, (int) (data.size() * 0.8));System.out.println(datass.x().instance(0));//訓練集System.out.println(datass.y().instance(0));//測試集Classifier c = new LibSVM();c.buildClassifier(datass.x());Map<Object,PerformanceMeasure> pms = EvaluateDataset.testDataset(c, datass.y());System.out.println(pms);/*for (int i = 0; i < 5; i++) {Pair<Dataset, Dataset> datas = s.sample(data, (int) (data.size() * 0.8), i);Classifier c = new LibSVM();c.buildClassifier(datas.x());Map<Object,PerformanceMeasure> pms = EvaluateDataset.testDataset(c, datas.y());System.out.println(pms);}*/} }

《新程序員》：云原生和全面數字化實踐50位技術專家共同創作，文字、視頻、音頻交互閱讀

總結

以上是生活随笔為你收集整理的Java机器学习库ML之三Sampling(采样)的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。