Question

public class Evaluate {

    public static void main(String args[]) throws Exception{
        //load datasets
        DataSource source = new DataSource("F:/data/solar_flare.arff");
        Instances dataset = source.getDataSet();
        BufferedReader reader = new BufferedReader(new FileReader("F:/data/solar_flare.arff"));
        Instances training = new Instances(reader);
        reader = new BufferedReader(new FileReader("F:/data/solar_flare.arff"));
        Instances testing = new Instances(reader);
        J48 jjjj = new J48();
        MultilabelClassifier PS = new PS();
        PS.setClassifier(jjjj);
        PS.setOptions(Utils.splitOptions("-threshold PCut1 -verbosity 3"));

        PS.buildClassifier(training);
        Evaluation ecal = new Evaluation(training);
        ecal.evaluateModel(PS, testing);

        dataset.setClassIndex(dataset.numAttributes()-1);
        J48 tree = new J48();
        tree.buildClassifier(dataset);
        Evaluation eval = new Evaluation(dataset);  
        Random rand = new Random(1);
        int folds = 10;
        DataSource source1 = new DataSource("F:/data/solar_flare.arff");
        Instances testDataset = source1.getDataSet();
        testDataset.setClassIndex(testDataset.numAttributes()-1);
        eval.crossValidateModel(tree, testDataset, folds, rand);
        System.out.println(eval.toMatrixString("=== Confusion Matrix ===\n"));

    }
}

这是我的代码，它给出了错误回复

这是错误

＆＃34;线程中的异常＆＃34; main＆＃34; java.lang.NoSuchMethodError： weka.core.Attribute。（Ljava / lang / String; Ljava / util / List;）V at meka.core.PSUtils.PSTransformation（PSUtils.java:416）at meka.classifiers.multilabel.PS.buildClassifier（PS.java:225）＆＃34;在 Evaluate.main（Evaluate.java:57）

Answer 1

当您使用的库的版本与您期望的版本不符时，这是一个典型的错误。您必须确保使用指定的方法在您的版本中存在weka.core.Attribute。

Meka api我想在Java Eclipse中集成Meka

1 个答案: