Example usage for org.apache.commons.math.linear RealVector getL1Norm

Introduction

In this page you can find the example usage for org.apache.commons.math.linear RealVector getL1Norm.

Prototype

double getL1Norm();

Source Link

Document

Returns the L₁ norm of the vector.

Usage

From source file:io.seldon.mf.RecentMfRecommender.java

public double[] createAvgProductVector(List<Long> recentitemInteractions, Map<Long, float[]> productFeatures) {
    int numLatentFactors = productFeatures.values().iterator().next().length;
    double[] userFeatures = new double[numLatentFactors];
    for (Long item : recentitemInteractions) {
        float[] productFactors = productFeatures.get(item);
        if (productFactors != null) {
            for (int feature = 0; feature < numLatentFactors; feature++) {
                userFeatures[feature] += productFactors[feature];
            }/*from   www  . ja  v  a  2s  .co m*/
        }
    }
    RealVector userFeaturesAsVector = new ArrayRealVector(userFeatures);
    RealVector normalised = userFeaturesAsVector.mapDivide(userFeaturesAsVector.getL1Norm());

    return normalised.getData();
}

---

From source file:com.opengamma.analytics.math.matrix.CommonsMatrixAlgebra.java

/**
 * {@inheritDoc}/*from   w  w w. j  a  va  2  s  .  com*/
 */
@Override
public double getNorm1(final Matrix<?> m) {
    Validate.notNull(m, "m");
    if (m instanceof DoubleMatrix1D) {
        final RealVector temp = CommonsMathWrapper.wrap((DoubleMatrix1D) m);
        return temp.getL1Norm();
    } else if (m instanceof DoubleMatrix2D) {
        final RealMatrix temp = CommonsMathWrapper.wrap((DoubleMatrix2D) m);
        // TODO find if commons implements this anywhere, so we are not doing it
        // by hand
        double max = 0.0;
        for (int col = temp.getColumnDimension() - 1; col >= 0; col--) {
            max = Math.max(max, temp.getColumnVector(col).getL1Norm());
        }
        return max;

    }
    throw new IllegalArgumentException("Can only find norm1 of DoubleMatrix2D; have " + m.getClass());
}

---

From source file:org.plista.kornakapi.core.recommender.FoldingFactorization.java

public double[] foldInAnonymousUser(long[] itemIDs) throws NoSuchItemException {

    double[] userFeatures = new double[factorization.numFeatures()];
    for (long itemID : itemIDs) {
        try {/*from w w  w.  j  a  va2s. c o  m*/
            int itemIndex = factorization.itemIndex(itemID);
            for (int feature = 0; feature < factorization.numFeatures(); feature++) {
                userFeatures[feature] += factorization.allItemFeatures()[itemIndex][feature];
            }
        } catch (NoSuchItemException e) {
            if (log.isDebugEnabled()) {
                log.debug("Item unknown: {}", itemID);
                if (itemIDs.length == 1) {
                    throw new NoSuchItemException("At least one item must be known");
                }
            }
        }
    }
    RealVector userFeaturesAsVector = new ArrayRealVector(userFeatures);
    RealVector normalised = userFeaturesAsVector.mapDivide(userFeaturesAsVector.getL1Norm());

    return normalised.getData();
}

---