Re: weka - failed to make a prediction

[Please reply to the list]

> //This one file must be in the "files" Directory (usually: Documents/JForex/files)
> @Library("C:/Users/rgmat/Documents/JForex/files/weka.jar")

This is only the core Weka library, not the AutoWeka (Weka) package.

>     public class WekaApp {
>
>         /** the model to use. */
>         protected Classifier cls;
>
>       /** the training data header. */
>       protected Instances header;
>
>       /**
>        * Load model and header.
>        *
>        * @throws Exception if deserialization fails
>        */
>        public void init() throws Exception {
>            // [A] load model and header of data
>             // deserialize model
>             Object[] objs = SerializationHelper.readAll("C:/TSFDC Trading 2022/JRip_EURNZD.model");
>             cls = (Classifier) objs[0];
>             header = (Instances) objs[1];
>        }
>
>        /**
>        * Returns the class label associated with the index.
>        *
>        * @param index the class label index
>        * @return the label
>        * @throws IllegalStateException if no header present
>        */
>        public String getLabel(int index) throws IllegalStateException {
>            if (header == null)
>                throw new IllegalStateException("No model/header loaded?");
>                return header.classAttribute().value(index);
>        }
>
>        /**
>        * Performs a prediction.
>        *
>        * @param v the value of the input variable to use for making a prediction
>        * @return the predicted value (class index), returns -1 if it fails to make prediction
>        * @throws Exception if prediction fails
>        */
>        public int currentInstance(double v) throws Exception {
>            try {
>                // [B] create instance
>                // create the current instance
>                double[] values = new double[header.numAttributes()];
>                for (int i = 0; i < values.length; i++)
>                    values[i] = Utils.missingValue();
>                    // - numeric
>                    values[0] = v;
>                    DenseInstance currentInstance = new DenseInstance(1.0, values);
>                    currentInstance.setDataset(header);
>
>                // [C] classify nominal class
>                // classifyInstance() just returns the index of the predicted label (the one with the highest probability) as a double
>                double prediction = cls.classifyInstance(currentInstance);
>                myConsole.getOut().println("predition: " + prediction);
>
>                return (int) prediction;
>            }
>            catch (Exception e) {
>                System.err.println("Weka exception occurred: " + e);
>                e.printStackTrace();
>                return -1;
>            }
>        }
>     } // end of class WekaApp

I don't see a main method that initializes the Weka packages, e.g. using:

weka.gui.GenericObjectEditor.determineClasses();


Cheers, Peter
--
Peter Reutemann
Dept. of Computer Science
University of Waikato, NZ
+64 (7) 858-5174 (office)
+64 (7) 577-5304 (home office)
https://www.cs.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