Examples with GLRMParameters hex.glrm.GLRMModel.GLRMParameters used on opensource projects

Example 11 with GLRMParameters

use of hex.glrm.GLRMModel.GLRMParameters in project h2o-3 by h2oai.

the class GLRMCategoricalTest method testCategoricalIris.

@Test
public void testCategoricalIris() throws InterruptedException, ExecutionException {
    GLRMModel model = null;
    Frame train = null;
    try {
        train = parse_test_file(Key.make("iris.hex"), "smalldata/iris/iris_wheader.csv");
        GLRMParameters parms = new GLRMParameters();
        parms._train = train._key;
        parms._k = 4;
        parms._loss = GlrmLoss.Absolute;
        parms._init = GlrmInitialization.SVD;
        parms._transform = DataInfo.TransformType.NONE;
        parms._recover_svd = true;
        parms._max_iterations = 1000;
        model = new GLRM(parms).trainModel().get();
        Log.info("Iteration " + model._output._iterations + ": Objective value = " + model._output._objective);
        model.score(train).delete();
        ModelMetricsGLRM mm = (ModelMetricsGLRM) ModelMetrics.getFromDKV(model, train);
        Log.info("Numeric Sum of Squared Error = " + mm._numerr + "\tCategorical Misclassification Error = " + mm._caterr);
    } finally {
        if (train != null)
            train.delete();
        if (model != null)
            model.delete();
    }
}

Example 12 with GLRMParameters

use of hex.glrm.GLRMModel.GLRMParameters in project h2o-3 by h2oai.

the class GLRMTest method testArrests.

@Test
public void testArrests() throws InterruptedException, ExecutionException {
    // Initialize using first k rows of standardized training frame
    Frame yinit = ArrayUtils.frame(ard(ard(1.24256408, 0.7828393, -0.5209066, -0.003416473), ard(0.50786248, 1.1068225, -1.2117642, 2.484202941), ard(0.07163341, 1.4788032, 0.9989801, 1.042878388)));
    GLRM job = null;
    GLRMModel model = null;
    Frame train = null;
    long seed = 1234;
    try {
        train = parse_test_file(Key.make("arrests.hex"), "smalldata/pca_test/USArrests.csv");
        GLRMParameters parms = new GLRMParameters();
        parms._train = train._key;
        parms._gamma_x = parms._gamma_y = 0.5;
        parms._regularization_x = GlrmRegularizer.Quadratic;
        parms._regularization_y = GlrmRegularizer.Quadratic;
        parms._k = 3;
        parms._transform = DataInfo.TransformType.STANDARDIZE;
        parms._init = GlrmInitialization.User;
        parms._recover_svd = false;
        parms._user_y = yinit._key;
        parms._seed = seed;
        job = new GLRM(parms);
        model = job.trainModel().get();
        Log.info("Iteration " + model._output._iterations + ": Objective value = " + model._output._objective);
        model.score(train).delete();
        ModelMetricsGLRM mm = (ModelMetricsGLRM) ModelMetrics.getFromDKV(model, train);
        Log.info("Numeric Sum of Squared Error = " + mm._numerr + "\tCategorical Misclassification Error = " + mm._caterr);
    } finally {
        yinit.delete();
        if (train != null)
            train.delete();
        if (model != null)
            model.delete();
    }
}

Example 13 with GLRMParameters

use of hex.glrm.GLRMModel.GLRMParameters in project h2o-3 by h2oai.

the class GLRMTest method testBenignSVD.

@Test
public void testBenignSVD() throws InterruptedException, ExecutionException {
    GLRM job = null;
    GLRMModel model = null;
    Frame train = null;
    try {
        train = parse_test_file(Key.make("benign.hex"), "smalldata/logreg/benign.csv");
        GLRMParameters parms = new GLRMParameters();
        parms._train = train._key;
        parms._k = 10;
        parms._gamma_x = parms._gamma_y = 0.25;
        parms._regularization_x = GlrmRegularizer.Quadratic;
        parms._regularization_y = GlrmRegularizer.Quadratic;
        parms._transform = DataInfo.TransformType.STANDARDIZE;
        parms._init = GlrmInitialization.SVD;
        parms._min_step_size = 1e-5;
        parms._recover_svd = false;
        parms._max_iterations = 2000;
        job = new GLRM(parms);
        model = job.trainModel().get();
        Log.info("Iteration " + model._output._iterations + ": Objective value = " + model._output._objective);
        model.score(train).delete();
        ModelMetricsGLRM mm = (ModelMetricsGLRM) ModelMetrics.getFromDKV(model, train);
        Log.info("Numeric Sum of Squared Error = " + mm._numerr + "\tCategorical Misclassification Error = " + mm._caterr);
    } finally {
        if (train != null)
            train.delete();
        if (model != null)
            model.delete();
    }
}

Example 14 with GLRMParameters

use of hex.glrm.GLRMModel.GLRMParameters in project h2o-3 by h2oai.

the class GLRMTest method testArrestsMissing.

@Test
public void testArrestsMissing() throws InterruptedException, ExecutionException {
    // Expected eigenvectors and their corresponding singular values with standardized data
    double[] sval = new double[] { 11.024148, 6.964086, 4.179904, 2.915146 };
    double[][] eigvec = ard(ard(-0.5358995, 0.4181809, -0.3412327, 0.64922780), ard(-0.5831836, 0.1879856, -0.2681484, -0.74340748), ard(-0.2781909, -0.8728062, -0.3780158, 0.13387773), ard(-0.5434321, -0.1673186, 0.8177779, 0.08902432));
    long seed = 1234;
    Frame train = null;
    GLRMModel model = null;
    GLRMParameters parms;
    Map<Double, Double> sd_map = new TreeMap<>();
    Map<Double, Double> ev_map = new TreeMap<>();
    StringBuilder sb = new StringBuilder();
    for (double missing_fraction : new double[] { 0, 0.1, 0.25, 0.5, 0.75, 0.9 }) {
        try {
            Scope.enter();
            train = parse_test_file(Key.make("arrests.hex"), "smalldata/pca_test/USArrests.csv");
            // Add missing values to the training data
            if (missing_fraction > 0) {
                Frame frtmp = new Frame(Key.<Frame>make(), train.names(), train.vecs());
                // Need to put the frame (to be modified) into DKV for MissingInserter to pick up
                DKV.put(frtmp._key, frtmp);
                FrameUtils.MissingInserter j = new FrameUtils.MissingInserter(frtmp._key, seed, missing_fraction);
                // MissingInserter is non-blocking, must block here explicitly
                j.execImpl().get();
                // Delete the frame header (not the data)
                DKV.remove(frtmp._key);
            }
            parms = new GLRMParameters();
            parms._train = train._key;
            parms._k = train.numCols();
            parms._loss = GlrmLoss.Quadratic;
            parms._regularization_x = GlrmRegularizer.None;
            parms._regularization_y = GlrmRegularizer.None;
            parms._transform = DataInfo.TransformType.STANDARDIZE;
            parms._init = GlrmInitialization.PlusPlus;
            parms._max_iterations = 1000;
            parms._seed = seed;
            parms._recover_svd = true;
            GLRM job = new GLRM(parms);
            model = job.trainModel().get();
            Log.info(100 * missing_fraction + "% missing values: Objective = " + model._output._objective);
            double sd_err = errStddev(sval, model._output._singular_vals) / parms._k;
            double ev_err = errEigvec(eigvec, model._output._eigenvectors_raw) / parms._k;
            Log.info("Avg SSE in Std Dev = " + sd_err + "\tAvg SSE in Eigenvectors = " + ev_err);
            sd_map.put(missing_fraction, sd_err);
            ev_map.put(missing_fraction, ev_err);
            model.score(train).delete();
            ModelMetricsGLRM mm = (ModelMetricsGLRM) ModelMetrics.getFromDKV(model, train);
            Log.info("Numeric Sum of Squared Error = " + mm._numerr + "\tCategorical Misclassification Error = " + mm._caterr);
            Assert.assertEquals(model._output._objective, mm._numerr, TOLERANCE);
            Scope.exit();
        } finally {
            if (train != null)
                train.delete();
            if (model != null)
                model.delete();
        }
    }
    sb.append("\nMissing Fraction --> Avg SSE in Std Dev\n");
    for (String s : Arrays.toString(sd_map.entrySet().toArray()).split(",")) sb.append(s.replace("=", " --> ")).append("\n");
    sb.append("\n");
    sb.append("Missing Fraction --> Avg SSE in Eigenvectors\n");
    for (String s : Arrays.toString(ev_map.entrySet().toArray()).split(",")) sb.append(s.replace("=", " --> ")).append("\n");
    Log.info(sb.toString());
}