Examples with DummyVectorizer org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer used on opensource projects

Example 11 with DummyVectorizer

use of org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer in project ignite by apache.

the class BinarizationExample method main.

/**
 * Run example.
 */
public static void main(String[] args) throws Exception {
    try (Ignite ignite = Ignition.start("examples/config/example-ignite.xml")) {
        System.out.println(">>> Binarization example started.");
        IgniteCache<Integer, Vector> data = null;
        try {
            data = createCache(ignite);
            Vectorizer<Integer, Vector, Integer, Double> vectorizer = new DummyVectorizer<>(1);
            // Defines second preprocessor that binarizes features.
            Preprocessor<Integer, Vector> preprocessor = new BinarizationTrainer<Integer, Vector>().withThreshold(40).fit(ignite, data, vectorizer);
            // Creates a cache based simple dataset containing features and providing standard dataset API.
            try (SimpleDataset<?> dataset = DatasetFactory.createSimpleDataset(ignite, data, preprocessor)) {
                new DatasetHelper(dataset).describe();
            }
            System.out.println(">>> Binarization example completed.");
        } finally {
            data.destroy();
        }
    } finally {
        System.out.flush();
    }
}

Example 12 with DummyVectorizer

use of org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer in project ignite by apache.

the class MaxAbsScalerExample method main.

/**
 * Run example.
 */
public static void main(String[] args) throws Exception {
    try (Ignite ignite = Ignition.start("examples/config/example-ignite.xml")) {
        System.out.println(">>> Max abs example started.");
        IgniteCache<Integer, Vector> data = null;
        try {
            data = createCache(ignite);
            Vectorizer<Integer, Vector, Integer, Double> vectorizer = new DummyVectorizer<>(1, 2);
            // Defines second preprocessor that imputing features.
            Preprocessor<Integer, Vector> preprocessor = new MaxAbsScalerTrainer<Integer, Vector>().fit(ignite, data, vectorizer);
            // Creates a cache based simple dataset containing features and providing standard dataset API.
            try (SimpleDataset<?> dataset = DatasetFactory.createSimpleDataset(ignite, data, preprocessor)) {
                new DatasetHelper(dataset).describe();
            }
            System.out.println(">>> Max abs example completed.");
        } finally {
            data.destroy();
        }
    } finally {
        System.out.flush();
    }
}

Example 13 with DummyVectorizer

use of org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer in project ignite by apache.

the class NormalizationExample method main.

/**
 * Run example.
 */
public static void main(String[] args) throws Exception {
    try (Ignite ignite = Ignition.start("examples/config/example-ignite.xml")) {
        System.out.println(">>> Normalization example started.");
        IgniteCache<Integer, Vector> data = null;
        try {
            data = createCache(ignite);
            Vectorizer<Integer, Vector, Integer, Double> vectorizer = new DummyVectorizer<>(1, 2);
            // Defines second preprocessor that normalizes features.
            Preprocessor<Integer, Vector> preprocessor = new NormalizationTrainer<Integer, Vector>().withP(1).fit(ignite, data, vectorizer);
            // Creates a cache based simple dataset containing features and providing standard dataset API.
            try (SimpleDataset<?> dataset = DatasetFactory.createSimpleDataset(ignite, data, preprocessor)) {
                new DatasetHelper(dataset).describe();
            }
            System.out.println(">>> Normalization example completed.");
        } finally {
            data.destroy();
        }
    } finally {
        System.out.flush();
    }
}

Example 14 with DummyVectorizer

use of org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer in project ignite by apache.

the class OneVsRestClassificationExample method main.

/**
 * Run example.
 */
public static void main(String[] args) throws IOException {
    System.out.println();
    System.out.println(">>> One-vs-Rest SVM Multi-class classification model over cached dataset usage example started.");
    // Start ignite grid.
    try (Ignite ignite = Ignition.start("examples/config/example-ignite.xml")) {
        System.out.println(">>> Ignite grid started.");
        IgniteCache<Integer, Vector> dataCache = null;
        try {
            dataCache = new SandboxMLCache(ignite).fillCacheWith(MLSandboxDatasets.GLASS_IDENTIFICATION);
            OneVsRestTrainer<SVMLinearClassificationModel> trainer = new OneVsRestTrainer<>(new SVMLinearClassificationTrainer().withAmountOfIterations(20).withAmountOfLocIterations(50).withLambda(0.2).withSeed(1234L));
            MultiClassModel<SVMLinearClassificationModel> mdl = trainer.fit(ignite, dataCache, new DummyVectorizer<Integer>().labeled(0));
            System.out.println(">>> One-vs-Rest SVM Multi-class model");
            System.out.println(mdl.toString());
            MinMaxScalerTrainer<Integer, Vector> minMaxScalerTrainer = new MinMaxScalerTrainer<>();
            Preprocessor<Integer, Vector> preprocessor = minMaxScalerTrainer.fit(ignite, dataCache, new DummyVectorizer<Integer>().labeled(0));
            MultiClassModel<SVMLinearClassificationModel> mdlWithScaling = trainer.fit(ignite, dataCache, preprocessor);
            System.out.println(">>> One-vs-Rest SVM Multi-class model with MinMaxScaling");
            System.out.println(mdlWithScaling.toString());
            System.out.println(">>> ----------------------------------------------------------------");
            System.out.println(">>> | Prediction\t| Prediction with MinMaxScaling\t| Ground Truth\t|");
            System.out.println(">>> ----------------------------------------------------------------");
            int amountOfErrors = 0;
            int amountOfErrorsWithMinMaxScaling = 0;
            int totalAmount = 0;
            // Build confusion matrix. See https://en.wikipedia.org/wiki/Confusion_matrix
            int[][] confusionMtx = { { 0, 0, 0 }, { 0, 0, 0 }, { 0, 0, 0 } };
            int[][] confusionMtxWithMinMaxScaling = { { 0, 0, 0 }, { 0, 0, 0 }, { 0, 0, 0 } };
            try (QueryCursor<Cache.Entry<Integer, Vector>> observations = dataCache.query(new ScanQuery<>())) {
                for (Cache.Entry<Integer, Vector> observation : observations) {
                    Vector val = observation.getValue();
                    Vector inputs = val.copyOfRange(1, val.size());
                    double groundTruth = val.get(0);
                    double prediction = mdl.predict(inputs);
                    double predictionWithMinMaxScaling = mdlWithScaling.predict(inputs);
                    totalAmount++;
                    // Collect data for model
                    if (!Precision.equals(groundTruth, prediction, Precision.EPSILON))
                        amountOfErrors++;
                    int idx1 = (int) prediction == 1 ? 0 : ((int) prediction == 3 ? 1 : 2);
                    int idx2 = (int) groundTruth == 1 ? 0 : ((int) groundTruth == 3 ? 1 : 2);
                    confusionMtx[idx1][idx2]++;
                    // Collect data for model with min-max scaling
                    if (!Precision.equals(groundTruth, predictionWithMinMaxScaling, Precision.EPSILON))
                        amountOfErrorsWithMinMaxScaling++;
                    idx1 = (int) predictionWithMinMaxScaling == 1 ? 0 : ((int) predictionWithMinMaxScaling == 3 ? 1 : 2);
                    idx2 = (int) groundTruth == 1 ? 0 : ((int) groundTruth == 3 ? 1 : 2);
                    confusionMtxWithMinMaxScaling[idx1][idx2]++;
                    System.out.printf(">>> | %.4f\t\t| %.4f\t\t\t\t\t\t| %.4f\t\t|\n", prediction, predictionWithMinMaxScaling, groundTruth);
                }
                System.out.println(">>> ----------------------------------------------------------------");
                System.out.println("\n>>> -----------------One-vs-Rest SVM model-------------");
                System.out.println("\n>>> Absolute amount of errors " + amountOfErrors);
                System.out.println("\n>>> Accuracy " + (1 - amountOfErrors / (double) totalAmount));
                System.out.println("\n>>> Confusion matrix is " + Arrays.deepToString(confusionMtx));
                System.out.println("\n>>> -----------------One-vs-Rest SVM model with MinMaxScaling-------------");
                System.out.println("\n>>> Absolute amount of errors " + amountOfErrorsWithMinMaxScaling);
                System.out.println("\n>>> Accuracy " + (1 - amountOfErrorsWithMinMaxScaling / (double) totalAmount));
                System.out.println("\n>>> Confusion matrix is " + Arrays.deepToString(confusionMtxWithMinMaxScaling));
                System.out.println(">>> One-vs-Rest SVM model over cache based dataset usage example completed.");
            }
        } finally {
            if (dataCache != null)
                dataCache.destroy();
        }
    } finally {
        System.out.flush();
    }
}

Example 15 with DummyVectorizer

use of org.apache.ignite.ml.dataset.feature.extractor.impl.DummyVectorizer in project ignite by apache.

the class LinearRegressionLSQRTrainerExample method main.

/**
 * Run example.
 */
public static void main(String[] args) throws IOException {
    System.out.println();
    System.out.println(">>> Linear regression model over cache based dataset usage example started.");
    // Start ignite grid.
    try (Ignite ignite = Ignition.start("examples/config/example-ignite.xml")) {
        System.out.println(">>> Ignite grid started.");
        IgniteCache<Integer, Vector> dataCache = null;
        try {
            dataCache = new SandboxMLCache(ignite).fillCacheWith(MLSandboxDatasets.MORTALITY_DATA);
            System.out.println(">>> Create new linear regression trainer object.");
            LinearRegressionLSQRTrainer trainer = new LinearRegressionLSQRTrainer();
            System.out.println(">>> Perform the training to get the model.");
            // This object is used to extract features and vectors from upstream entities which are
            // essentially tuples of the form (key, value) (in our case (Integer, Vector)).
            // Key part of tuple in our example is ignored.
            // Label is extracted from 0th entry of the value (which is a Vector)
            // and features are all remaining vector part. Alternatively we could use
            // DatasetTrainer#fit(Ignite, IgniteCache, IgniteBiFunction, IgniteBiFunction) method call
            // where there is a separate lambda for extracting label from (key, value) and a separate labmda for
            // extracting features.
            LinearRegressionModel mdl = trainer.fit(ignite, dataCache, new DummyVectorizer<Integer>().labeled(Vectorizer.LabelCoordinate.FIRST));
            double rmse = Evaluator.evaluate(dataCache, mdl, new DummyVectorizer<Integer>().labeled(Vectorizer.LabelCoordinate.FIRST), MetricName.RMSE);
            System.out.println("\n>>> Rmse = " + rmse);
            System.out.println(">>> Linear regression model over cache based dataset usage example completed.");
        } finally {
            if (dataCache != null)
                dataCache.destroy();
        }
    } finally {
        System.out.flush();
    }
}