Example usage for weka.classifiers.timeseries WekaForecaster setFieldsToForecast

List of usage examples for weka.classifiers.timeseries WekaForecaster setFieldsToForecast

Introduction

In this page you can find the example usage for weka.classifiers.timeseries WekaForecaster setFieldsToForecast.

Prototype

@Override
public void setFieldsToForecast(String fieldsToForecast) throws Exception 

Source Link

Document

Set the names of the fields/attributes in the data to forecast.

Usage

From source file:org.datahack.forecast.BayExample.java

public static void main(String[] args) {
    try {//from   ww  w  .j  a  v a  2  s.  c om
        // path to the Australian wine data included with the time series forecasting
        // package

        File output = File.createTempFile("tempWineData", "arff");
        output.deleteOnExit();
        String dataFileName = "sample-data/wine.arff";
        URL resource = BayExample.class.getResource("/" + dataFileName);
        FileUtils.copyURLToFile(resource, output);

        String pathToWineData = output.getPath();

        // load the wine data
        Instances wine = new Instances(new BufferedReader(new FileReader(pathToWineData)));

        InstanceQuery q = new InstanceQuery();

        // new forecaster
        WekaForecaster forecaster = new WekaForecaster();

        // set the targets we want to forecast. This method calls
        // setFieldsToLag() on the lag maker object for us
        forecaster.setFieldsToForecast("Fortified,Dry-white");

        // default underlying classifier is SMOreg (SVM) - we'll use
        // gaussian processes for regression instead
        forecaster.setBaseForecaster(new GaussianProcesses());

        forecaster.getTSLagMaker().setTimeStampField("Date"); // date time stamp
        forecaster.getTSLagMaker().setMinLag(1);
        forecaster.getTSLagMaker().setMaxLag(12); // monthly data

        // add a month of the year indicator field
        forecaster.getTSLagMaker().setAddMonthOfYear(true);

        // add a quarter of the year indicator field
        forecaster.getTSLagMaker().setAddQuarterOfYear(true);

        // build the model
        forecaster.buildForecaster(wine, System.out);

        // prime the forecaster with enough recent historical data
        // to cover up to the maximum lag. In our case, we could just supply
        // the 12 most recent historical instances, as this covers our maximum
        // lag period
        forecaster.primeForecaster(wine);

        // forecast for 12 units (months) beyond the end of the
        // training data
        List<List<NumericPrediction>> forecast = forecaster.forecast(12, System.out);

        // output the predictions. Outer list is over the steps; inner list is over
        // the targets
        for (int i = 0; i < 12; i++) {
            List<NumericPrediction> predsAtStep = forecast.get(i);
            for (int j = 0; j < 2; j++) {
                NumericPrediction predForTarget = predsAtStep.get(j);
                System.out.print("" + predForTarget.predicted() + " ");
            }
            System.out.println();
        }

        // we can continue to use the trained forecaster for further forecasting
        // by priming with the most recent historical data (as it becomes available).
        // At some stage it becomes prudent to re-build the model using current
        // historical data.

    } catch (Exception ex) {
        ex.printStackTrace();
    }
}

From source file:org.datahack.forecast.TimeSeriesExample.java

public static void main(String[] args) {
    try {/*w  ww . ja v a  2s.c  om*/
        // path to the Australian wine data included with the time series forecasting
        // package

        File output = File.createTempFile("tempWineData", "arff");
        output.deleteOnExit();
        String dataFileName = "sample-data/parking344.arff";
        URL resource = TimeSeriesExample.class.getResource("/" + dataFileName);
        FileUtils.copyURLToFile(resource, output);

        String pathToWineData = output.getPath();

        // load the wine data
        Instances wine = new Instances(new BufferedReader(new FileReader(pathToWineData)));

        // new forecaster
        WekaForecaster forecaster = new WekaForecaster();

        // set the targets we want to forecast. This method calls
        // setFieldsToLag() on the lag maker object for us
        forecaster.setFieldsToForecast("occupiedSpaces");

        // default underlying classifier is SMOreg (SVM) - we'll use
        // gaussian processes for regression instead
        forecaster.setBaseForecaster(new GaussianProcesses());

        forecaster.getTSLagMaker().setTimeStampField("eventTime"); // date time stamp
        forecaster.getTSLagMaker().setMinLag(1);
        forecaster.getTSLagMaker().setMaxLag(12); // monthly data

        // add a month of the year indicator field
        forecaster.getTSLagMaker().setAddMonthOfYear(true);

        // add a quarter of the year indicator field
        forecaster.getTSLagMaker().setAddQuarterOfYear(true);

        // build the model
        forecaster.buildForecaster(wine, System.out);

        // prime the forecaster with enough recent historical data
        // to cover up to the maximum lag. In our case, we could just supply
        // the 12 most recent historical instances, as this covers our maximum
        // lag period
        forecaster.primeForecaster(wine);

        // forecast for 12 units (months) beyond the end of the
        // training data
        List<List<NumericPrediction>> forecast = forecaster.forecast(12, System.out);

        // output the predictions. Outer list is over the steps; inner list is over
        // the targets
        for (int i = 0; i < 12; i++) {
            List<NumericPrediction> predsAtStep = forecast.get(i);
            for (int j = 0; j < 2; j++) {
                NumericPrediction predForTarget = predsAtStep.get(j);
                System.out.print("" + predForTarget.predicted() + " ");
            }
            System.out.println();
        }

        // we can continue to use the trained forecaster for further forecasting
        // by priming with the most recent historical data (as it becomes available).
        // At some stage it becomes prudent to re-build the model using current
        // historical data.

    } catch (Exception ex) {
        ex.printStackTrace();
    }
}

From source file:Prediccion.Prediccion.java

License:Open Source License

@Override
public void run() {
    try {/*from w w w  .j a  v a 2 s.  c o m*/

        ArrayList<Instances> pasos = cargarDatos();

        System.err.println(pasos.size());

        //Instanciamos el predictor
        ArrayList<WekaForecaster> forecaster = new ArrayList<>(24);

        for (int a = 0; a < 24; a++) {
            forecaster.add(new WekaForecaster());
        }

        int a = 0;

        for (WekaForecaster fore : forecaster) {

            //Defimimos el atributo que queremos predecir
            fore.setFieldsToForecast("Total");

            //Definimos el mtodo de prediccin a emplear. En este caso, regresin lineal porque 
            //en el artculo es el que mejor ha funcionado
            fore.setBaseForecaster(new LinearRegression());

            //Defimimos el atributo que "marca" el tiempo y su peridiocidad
            fore.getTSLagMaker().setTimeStampField("Intervalo");
            fore.getTSLagMaker().setMinLag(1);
            fore.getTSLagMaker().setMaxLag(1);

            fore.getTSLagMaker().setPeriodicity(TSLagMaker.Periodicity.WEEKLY);

            fore.buildForecaster(pasos.get(a), System.out);

            // System.err.println(pasos.get(a).toString());

            //System.err.printf("Termin");

            fore.primeForecaster(pasos.get(a));

            List<List<NumericPrediction>> forecast = fore.forecast(1, System.out);

            System.err.println("==== " + a + " ====");
            // output the predictions. Outer list is over the steps; inner list is over
            // the targets
            for (int i = 0; i < 1; i++) {
                List<NumericPrediction> predsAtStep = forecast.get(i);
                for (int j = 0; j < 1; j++) {
                    NumericPrediction predForTarget = predsAtStep.get(j);
                    System.err.print("" + predForTarget.predicted() + " ");
                }
                System.err.println();
            }
            a++;
        }
        /*    
                    
           // path to the Australian wine data included with the time series forecasting
           // package
           String pathToWineData = weka.core.WekaPackageManager.PACKAGES_DIR.toString()
             + File.separator + "timeseriesForecasting" + File.separator + "sample-data"
             + File.separator + "wine.arff";
                
           // load the wine data
           Instances wine = new Instances(new BufferedReader(new FileReader(pathToWineData)));      
                   
           // new forecaster
           WekaForecaster forecaster = new WekaForecaster();
                
           // set the targets we want to forecast. This method calls
           // setFieldsToLag() on the lag maker object for us
           forecaster.setFieldsToForecast("Fortified,Dry-white");
                
           // default underlying classifier is SMOreg (SVM) - we'll use
           // gaussian processes for regression instead
           forecaster.setBaseForecaster(new GaussianProcesses());
                
           forecaster.getTSLagMaker().setTimeStampField("Date"); // date time stamp
           forecaster.getTSLagMaker().setMinLag(1);
           forecaster.getTSLagMaker().setMaxLag(12); // monthly data
                
           // add a month of the year indicator field
           forecaster.getTSLagMaker().setAddMonthOfYear(true);
                
           // add a quarter of the year indicator field
           forecaster.getTSLagMaker().setAddQuarterOfYear(true);
                
           // build the model
           forecaster.buildForecaster(wine, System.out);
                
           // prime the forecaster with enough recent historical data
           // to cover up to the maximum lag. In our case, we could just supply
           // the 12 most recent historical instances, as this covers our maximum
           // lag period
           forecaster.primeForecaster(wine);
                
           // forecast for 12 units (months) beyond the end of the
           // training data
           <<List<List<NumericPrediction>> forecast = forecaster.forecast(12, System.out);
                
                   
                   
           // output the predictions. Outer list is over the steps; inner list is over
           // the targets
           for (int i = 0; i < 12; i++) {
             List<NumericPrediction> predsAtStep = forecast.get(i);
             for (int j = 0; j < 2; j++) {
               NumericPrediction predForTarget = predsAtStep.get(j);
               System.out.print("" + predForTarget.predicted() + " ");
             }
             System.out.println();
           }
                
           // we can continue to use the trained forecaster for further forecasting
           // by priming with the most recent historical data (as it becomes available).
           // At some stage it becomes prudent to re-build the model using current
           // historical data.
        */
    } catch (Exception ex) {
        ex.printStackTrace();
    }

}