Examples with MLMeasures edu.neu.ccs.pyramid.eval.MLMeasures used on opensource projects

Example 1 with MLMeasures

use of edu.neu.ccs.pyramid.eval.MLMeasures in project pyramid by cheng-li.

the class CBMEN method reportAccPrediction.

private static void reportAccPrediction(Config config, CBM cbm, MultiLabelClfDataSet dataSet) throws Exception {
    System.out.println("============================================================");
    System.out.println("Making predictions on test set with the instance set accuracy optimal predictor");
    String output = config.getString("output.dir");
    AccPredictor accPredictor = new AccPredictor(cbm);
    accPredictor.setComponentContributionThreshold(config.getDouble("predict.piThreshold"));
    MultiLabel[] predictions = accPredictor.predict(dataSet);
    MLMeasures mlMeasures = new MLMeasures(dataSet.getNumClasses(), dataSet.getMultiLabels(), predictions);
    System.out.println("test performance with the instance set accuracy optimal predictor");
    System.out.println(mlMeasures);
    File performanceFile = Paths.get(output, "test_predictions", "instance_accuracy_optimal", "performance.txt").toFile();
    FileUtils.writeStringToFile(performanceFile, mlMeasures.toString());
    System.out.println("test performance is saved to " + performanceFile.toString());
    // Here we do not use approximation
    double[] setProbs = IntStream.range(0, predictions.length).parallel().mapToDouble(i -> cbm.predictAssignmentProb(dataSet.getRow(i), predictions[i])).toArray();
    File predictionFile = Paths.get(output, "test_predictions", "instance_accuracy_optimal", "predictions.txt").toFile();
    try (BufferedWriter br = new BufferedWriter(new FileWriter(predictionFile))) {
        for (int i = 0; i < dataSet.getNumDataPoints(); i++) {
            br.write(predictions[i].toString());
            br.write(":");
            br.write("" + setProbs[i]);
            br.newLine();
        }
    }
    System.out.println("predicted sets and their probabilities are saved to " + predictionFile.getAbsolutePath());
    System.out.println("============================================================");
}

Example 2 with MLMeasures

use of edu.neu.ccs.pyramid.eval.MLMeasures in project pyramid by cheng-li.

the class CBMLR method tune.

private static TuneResult tune(Config config, HyperParameters hyperParameters, MultiLabelClfDataSet trainSet, MultiLabelClfDataSet validSet) throws Exception {
    CBM cbm = newCBM(config, trainSet, hyperParameters);
    EarlyStopper earlyStopper = loadNewEarlyStopper(config);
    LRCBMOptimizer optimizer = getOptimizer(config, hyperParameters, cbm, trainSet);
    optimizer.initialize();
    MultiLabelClassifier classifier;
    String predictTarget = config.getString("tune.targetMetric");
    switch(predictTarget) {
        case "instance_set_accuracy":
            AccPredictor accPredictor = new AccPredictor(cbm);
            accPredictor.setComponentContributionThreshold(config.getDouble("predict.piThreshold"));
            classifier = accPredictor;
            break;
        case "instance_f1":
            PluginF1 pluginF1 = new PluginF1(cbm);
            List<MultiLabel> support = DataSetUtil.gatherMultiLabels(trainSet);
            pluginF1.setSupport(support);
            pluginF1.setPiThreshold(config.getDouble("predict.piThreshold"));
            classifier = pluginF1;
            break;
        case "instance_hamming_loss":
            MarginalPredictor marginalPredictor = new MarginalPredictor(cbm);
            marginalPredictor.setPiThreshold(config.getDouble("predict.piThreshold"));
            classifier = marginalPredictor;
            break;
        default:
            throw new IllegalArgumentException("predictTarget should be instance_set_accuracy, instance_f1 or instance_hamming_loss");
    }
    int interval = config.getInt("tune.monitorInterval");
    for (int iter = 1; true; iter++) {
        if (VERBOSE) {
            System.out.println("iteration " + iter);
        }
        optimizer.iterate();
        if (iter % interval == 0) {
            MLMeasures validMeasures = new MLMeasures(classifier, validSet);
            if (VERBOSE) {
                System.out.println("validation performance with " + predictTarget + " optimal predictor:");
                System.out.println(validMeasures);
            }
            switch(predictTarget) {
                case "instance_set_accuracy":
                    earlyStopper.add(iter, validMeasures.getInstanceAverage().getAccuracy());
                    break;
                case "instance_f1":
                    earlyStopper.add(iter, validMeasures.getInstanceAverage().getF1());
                    break;
                case "instance_hamming_loss":
                    earlyStopper.add(iter, validMeasures.getInstanceAverage().getHammingLoss());
                    break;
                default:
                    throw new IllegalArgumentException("predictTarget should be instance_set_accuracy or instance_f1");
            }
            if (earlyStopper.shouldStop()) {
                if (VERBOSE) {
                    System.out.println("Early Stopper: the training should stop now!");
                }
                break;
            }
        }
    }
    if (VERBOSE) {
        System.out.println("done!");
    }
    hyperParameters.iterations = earlyStopper.getBestIteration();
    TuneResult tuneResult = new TuneResult();
    tuneResult.hyperParameters = hyperParameters;
    tuneResult.performance = earlyStopper.getBestValue();
    return tuneResult;
}

Example 3 with MLMeasures

use of edu.neu.ccs.pyramid.eval.MLMeasures in project pyramid by cheng-li.

the class CBMLR method reportHammingPrediction.

private static void reportHammingPrediction(Config config, CBM cbm, MultiLabelClfDataSet dataSet) throws Exception {
    System.out.println("============================================================");
    System.out.println("Making predictions on test set with the instance Hamming loss optimal predictor");
    String output = config.getString("output.dir");
    MarginalPredictor marginalPredictor = new MarginalPredictor(cbm);
    marginalPredictor.setPiThreshold(config.getDouble("predict.piThreshold"));
    MultiLabel[] predictions = marginalPredictor.predict(dataSet);
    MLMeasures mlMeasures = new MLMeasures(dataSet.getNumClasses(), dataSet.getMultiLabels(), predictions);
    System.out.println("test performance with the instance Hamming loss optimal predictor");
    System.out.println(mlMeasures);
    File performanceFile = Paths.get(output, "test_predictions", "instance_hamming_loss_optimal", "performance.txt").toFile();
    FileUtils.writeStringToFile(performanceFile, mlMeasures.toString());
    System.out.println("test performance is saved to " + performanceFile.toString());
    // Here we do not use approximation
    double[] setProbs = IntStream.range(0, predictions.length).parallel().mapToDouble(i -> cbm.predictAssignmentProb(dataSet.getRow(i), predictions[i])).toArray();
    File predictionFile = Paths.get(output, "test_predictions", "instance_hamming_loss_optimal", "predictions.txt").toFile();
    try (BufferedWriter br = new BufferedWriter(new FileWriter(predictionFile))) {
        for (int i = 0; i < dataSet.getNumDataPoints(); i++) {
            br.write(predictions[i].toString());
            br.write(":");
            br.write("" + setProbs[i]);
            br.newLine();
        }
    }
    System.out.println("predicted sets and their probabilities are saved to " + predictionFile.getAbsolutePath());
    System.out.println("============================================================");
}

Example 4 with MLMeasures

use of edu.neu.ccs.pyramid.eval.MLMeasures in project pyramid by cheng-li.

the class CBMLR method reportAccPrediction.

private static void reportAccPrediction(Config config, CBM cbm, MultiLabelClfDataSet dataSet) throws Exception {
    System.out.println("============================================================");
    System.out.println("Making predictions on test set with the instance set accuracy optimal predictor");
    String output = config.getString("output.dir");
    AccPredictor accPredictor = new AccPredictor(cbm);
    accPredictor.setComponentContributionThreshold(config.getDouble("predict.piThreshold"));
    MultiLabel[] predictions = accPredictor.predict(dataSet);
    MLMeasures mlMeasures = new MLMeasures(dataSet.getNumClasses(), dataSet.getMultiLabels(), predictions);
    System.out.println("test performance with the instance set accuracy optimal predictor");
    System.out.println(mlMeasures);
    File performanceFile = Paths.get(output, "test_predictions", "instance_accuracy_optimal", "performance.txt").toFile();
    FileUtils.writeStringToFile(performanceFile, mlMeasures.toString());
    System.out.println("test performance is saved to " + performanceFile.toString());
    // Here we do not use approximation
    double[] setProbs = IntStream.range(0, predictions.length).parallel().mapToDouble(i -> cbm.predictAssignmentProb(dataSet.getRow(i), predictions[i])).toArray();
    File predictionFile = Paths.get(output, "test_predictions", "instance_accuracy_optimal", "predictions.txt").toFile();
    try (BufferedWriter br = new BufferedWriter(new FileWriter(predictionFile))) {
        for (int i = 0; i < dataSet.getNumDataPoints(); i++) {
            br.write(predictions[i].toString());
            br.write(":");
            br.write("" + setProbs[i]);
            br.newLine();
        }
    }
    System.out.println("predicted sets and their probabilities are saved to " + predictionFile.getAbsolutePath());
    System.out.println("============================================================");
}

Example 5 with MLMeasures

use of edu.neu.ccs.pyramid.eval.MLMeasures in project pyramid by cheng-li.

the class CBMGB method reportHammingPrediction.

private static void reportHammingPrediction(Config config, CBM cbm, MultiLabelClfDataSet dataSet) throws Exception {
    System.out.println("============================================================");
    System.out.println("Making predictions on test set with the instance Hamming loss optimal predictor");
    String output = config.getString("output.dir");
    MarginalPredictor marginalPredictor = new MarginalPredictor(cbm);
    marginalPredictor.setPiThreshold(config.getDouble("predict.piThreshold"));
    MultiLabel[] predictions = marginalPredictor.predict(dataSet);
    MLMeasures mlMeasures = new MLMeasures(dataSet.getNumClasses(), dataSet.getMultiLabels(), predictions);
    System.out.println("test performance with the instance Hamming loss optimal predictor");
    System.out.println(mlMeasures);
    File performanceFile = Paths.get(output, "test_predictions", "instance_hamming_loss_optimal", "performance.txt").toFile();
    FileUtils.writeStringToFile(performanceFile, mlMeasures.toString());
    System.out.println("test performance is saved to " + performanceFile.toString());
    // Here we do not use approximation
    double[] setProbs = IntStream.range(0, predictions.length).parallel().mapToDouble(i -> cbm.predictAssignmentProb(dataSet.getRow(i), predictions[i])).toArray();
    File predictionFile = Paths.get(output, "test_predictions", "instance_hamming_loss_optimal", "predictions.txt").toFile();
    try (BufferedWriter br = new BufferedWriter(new FileWriter(predictionFile))) {
        for (int i = 0; i < dataSet.getNumDataPoints(); i++) {
            br.write(predictions[i].toString());
            br.write(":");
            br.write("" + setProbs[i]);
            br.newLine();
        }
    }
    System.out.println("predicted sets and their probabilities are saved to " + predictionFile.getAbsolutePath());
    System.out.println("============================================================");
}