Re: ArrayIndexOutOfBoundsException when classifying an instance

Peter Reutemann <[email protected]> Wed, 19 May 2021 20:01:26 +1200

Newsgroups	gmane.comp.ai.weka
Message-ID	<CAHoQ12+nZD5xCRoNzL03VKseUMXQUZkMemV8ExCNh9x-jRvDAg@mail.gmail.com>

> I am trying to classify instances and code is given below.
>
> public void prepare_data() throws IOException {
>                 BufferedReader buf_reader=new BufferedReader(new
> FileReader(this.train_file));
>
>                 Instances train_data = new Instances(buf_reader);
>                 train_data.setClassIndex(train_data.numAttributes() - 1);
>                 this.train=train_data;
>                 buf_reader.close();
>
>                 buf_reader=new BufferedReader(new FileReader(this.test_file));
>
>                 Instances test_data = new Instances(buf_reader);
>                 test_data.setClassIndex(test_data.numAttributes() - 1);
>                 this.test=test_data;
>                 buf_reader.close();
>         }
>
> public void filteredFier() throws Exception {
>
>                 System.out.println("hi"+ this.train.equalHeaders(this.test));
>                 NumericToNominal nn = new NumericToNominal();
>                 String[] options= {"-R", "first-last"};
>                 nn.setOptions(options);
>                 nn.setInputFormat(this.train);
>                 this.rf=new RandomForest();
>                 this.fc = new FilteredClassifier();
>                 this.fc.setFilter(nn);
>
>                 this.fc.setClassifier(this.rf);
>
>                 this.fc.buildClassifier(this.train);
>
>                 Evaluation eval=new Evaluation(this.train);
>
>
>                 eval.crossValidateModel(this.fc,this.train,10,new Random(1));
>
>                 System.out.println(eval.toSummaryString());
>
>
>
>
>         }
>         public void get_accuracy() throws Exception {
>
>
>
>                 for (int i = 0; i < this.test.numInstances(); i++) {
>                         double clsLabel = this.fc.classifyInstance(this.test.instance(i));
>                         System.out.println(clsLabel);
>                 }
>
>
>
>
>         }
>
> I get ArrayIndexOutOfBoundException. Somebody here knows solution, I need
> your help. Thank you in advance.

You need to post the full stacktrace. With the information that you
provided, we don't know where the error originated.

I took your code and stream-lined it a bit. Run it with your
train/test set and post the full stacktrace if you still get the
ArrayIndexOutOfBoundsException.
The class expects two command-line arguments: train and test file.

Cheers, Peter
-- 
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 577-5304
http://www.cms.waikato.ac.nz/~fracpete/
http://www.data-mining.co.nz/

_______________________________________________
Wekalist mailing list -- [email protected]
Send posts to [email protected]
To unsubscribe send an email to [email protected]
To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

Something.java (text/x-java, 2.8 KB)

import weka.classifiers.AbstractClassifier;
import weka.classifiers.Classifier;
import weka.classifiers.Evaluation;
import weka.classifiers.evaluation.Prediction;
import weka.classifiers.meta.FilteredClassifier;
import weka.classifiers.trees.RandomForest;
import weka.core.Attribute;
import weka.core.Instances;
import weka.core.converters.ConverterUtils.DataSource;
import weka.filters.unsupervised.attribute.NumericToNominal;

import java.util.Random;

public class Something {

  private String trainFile;
  private Instances train;
  private String testFile;
  private Instances test;
  private FilteredClassifier fc;

  public Something(String trainFile, String testFile) {
    this.trainFile = trainFile;
    this.testFile = testFile;
  }

  public void loadData() throws Exception {
    train = DataSource.read(trainFile);
    train.setClassIndex(train.numAttributes() - 1);

    test = DataSource.read(testFile);
    test.setClassIndex(test.numAttributes() - 1);

    System.out.println("train/test compatible? " + train.equalHeaders(test));
  }

  public Instances getTrain() {
    return train;
  }

  public Instances getTest() {
    return test;
  }

  public void setupClassifier() throws Exception {
    NumericToNominal nn = new NumericToNominal();
    String[] options = {"-R", "first-last"};
    nn.setOptions(options);
    nn.setInputFormat(train);
    RandomForest rf = new RandomForest();
    fc = new FilteredClassifier();
    fc.setFilter(nn);
    fc.setClassifier(rf);
  }

  public Evaluation crossValidate() throws Exception {
    Evaluation eval = new Evaluation(train);
    eval.crossValidateModel(fc, train, 10, new Random(1));
    System.out.println(eval.toSummaryString("\n=== Cross-validation on train ===", false));
    return eval;
  }

  public Evaluation evaluate() throws Exception {
    Classifier cls = AbstractClassifier.makeCopy(fc);
    cls.buildClassifier(train);
    Evaluation eval = new Evaluation(train);
    eval.evaluateModel(cls, test);
    System.out.println(eval.toSummaryString("\n=== Evaluation on test ===", false));
    return eval;
  }

  public static void main(String[] args) throws Exception {
    Evaluation eval;
    Something s = new Something(args[0], args[1]);
    s.loadData();
    s.setupClassifier();

    // cross-validate on train
    eval = s.crossValidate();
    System.out.println("Accuracy (cross-validation): "  + eval.pctCorrect());

    // build on train and evaluate on test
    eval = s.evaluate();
    System.out.println("Accuracy (test): "  + eval.pctCorrect());

    // output predictions on test
    System.out.println("\nPredictions (actual -> predicted)");
    Attribute clsAtt = s.getTrain().classAttribute();
    for (Prediction p: eval.predictions()) {
      System.out.println(clsAtt.value((int) p.actual()) + " -> " + clsAtt.value((int) p.predicted()));
    }
  }
}