ERROR While using WEKA API in java code: Class Attribute Not Set?

Angga Raditya picture Angga Raditya · Nov 13, 2011 · Viewed 15.8k times · Source

I'm trying to use weka API in my java code. I use J48 tree classification to Classify my dataset in MySQL database, but I have this error:

Trying to add database driver (JDBC): RmiJdbc.RJDriver - Error, not in CLASSPATH?
Trying to add database driver (JDBC): jdbc.idbDriver - Error, not in CLASSPATH?
Trying to add database driver (JDBC): com.mckoi.JDBCDriver - Error, not in CLASSPATH?
Trying to add database driver (JDBC): org.hsqldb.jdbcDriver - Error, not in CLASSPATH?
weka.core.UnassignedClassException: weka.classifiers.trees.j48.C45PruneableClassifierTree: Class attribute not set!
        at weka.core.Capabilities.test(Capabilities.java:1086)
        at weka.core.Capabilities.test(Capabilities.java:1018)
        at weka.core.Capabilities.testWithFail(Capabilities.java:1297)
.....

This is my code:

try{
       DatabaseLoader loader = new DatabaseLoader();
      loader.setSource("jdbc:mysql://localhost:3306/cuaca","root","491754");
       loader.setQuery("select * from data_training");
      Instances data = loader.getDataSet();

        jTextArea1.append(data.toString());

        String[] options = new String[1];
        options[0] = "U"; // unpruned tree
        J48 tree = new J48(); // new instance of tree
       tree.setOptions(options); // set the option
       tree.buildClassifier(data); // build classifier
       //Evaluation eval=new Evaluation(data);
       //eval.crossValidateModel(tree, data, 10, new Random(1));
       //jTextArea1.append(eval.toSummaryString("\n HASILNYA \n", false));

 }catch(IOException ioe){ioe.printStackTrace();}
        catch(Exception e){e.printStackTrace();}

The dataset actually appears in the textarea:

@relation data_training

@attribute Tmean numeric
@attribute Tmax numeric
@attribute Td numeric
@attribute RH numeric
@attribute SLP numeric
@attribute STP numeric
@attribute Wind_ave numeric
@attribute Wind_max numeric
@attribute Hujan {Y,T}

@data
25.9,31.6,23.1,93.4,1008.5,998.2,2.6,12.9,Y
27.6,31.4,22.3,87,1009.6,999.3,0.8,3.1,T
27.4,32.6,21.9,86.1,1009.4,999.2,4.5,14.3,T
27.6,32.2,22.4,87.3,1009.1,998.9,2.4,8.2,T

I would to know why the error appear(Class Attribute Not Set!!)

Answer

iinception picture iinception · Nov 17, 2011

You should set class index of for your dataset before passing it into classifier. Your classifier must know which is your outcome variable.
After these lines...

loader.setQuery("select * from data_training");
Instances data = loader.getDataSet();

Add the following:

data.setClassIndex(data.numAttributes() - 1);

If Hujan is your class attribute (outcome variable)

See Api Docs for more Info