Examples of SymmetricMatrix

ch.akuhn.matrix.SymmetricMatrix
Matrix where a_ij = a_ji for all elements.
@author Adrian Kuhn
edu.ucla.sspace.matrix.SymmetricMatrix
A symmetric dense matrix that only stores the values of the lower triangular. This class only records changes values where row > col. For all other values, the row and column values are swapped and then the backing matrix is updated.
The primary benfit of this class is for storing large symmetric matrices in half of the memory. @author David Jurgens

Examples of ch.akuhn.matrix.SymmetricMatrix

    }
    public Mds(DenseMatrix dissimilarities, Points initial,  
            Function fConfigDist, Function fWeights, Function fDtarget) {
        len = dissimilarities.rowCount();
        Dtarget = dissimilarities;
        config_dist = new SymmetricMatrix(len);
        this.pos = initial == null ? new Points(len) : initial;
        this.gradient = new Points(len);
        f_config_dist = fConfigDist;
        f_weights = fWeights;
        f_dtarget = fDtarget;

View Full Code Here

Examples of ch.akuhn.matrix.SymmetricMatrix


        //double this_weight;
        //double local_weight_power = 0.;
        //double local_within_between = 1.;


        this.weights = new SymmetricMatrix(Dtarget.rowCount());
        for (int i = 0; i < Dtarget.rowCount(); i++) {
            for (int j = 0; j < i; j++) {
                this.weights.put(i,j, f_weights.apply(this.Dtarget.get(i,j)));
            }
        }

View Full Code Here

Examples of ch.akuhn.matrix.SymmetricMatrix

            constructCloserThanEpsilonNeighborhoodGraph();
        }
    }
    
    private void constructCloserThanEpsilonNeighborhoodGraph() {
        graph = new SymmetricMatrix(n);
        graph.fill(Double.POSITIVE_INFINITY);
        for (int i = 0; i < n; i++) {
            for (int j = 0; j < i; j++) {
                if (dist(i,j) < e) graph.put(i, j, e);
            }

View Full Code Here

Examples of ch.akuhn.matrix.SymmetricMatrix

            }
        }
    }
    
    private void constructKayNearestNeighborhoodGraph() {
        graph = new SymmetricMatrix(n);
        graph.fill(Double.POSITIVE_INFINITY);
        for (int i = 0; i < n; i++) {
            double[] data = new double[n];
            for (int j = 0; j < n; j++) data[j] = dist(i,j);
            int[] index = indicesOfMinima(data, k+1);

View Full Code Here

Examples of edu.ucla.sspace.matrix.SymmetricMatrix

     * SimilarityFunction} will be used to build a symmetric adjacency matrix
     * and the {@link ClusterLink} will determine how to udpate similarities
     * between newly nerged clusters.
     */
    public Assignments cluster(Matrix m, int numClusters, Properties props) {
        Matrix adj = new SymmetricMatrix(m.rows(), m.rows());
        for (int r = 0; r < m.rows(); ++r) {
            DoubleVector v = m.getRowVector(r);
            for (int c = r+1; c < m.rows(); ++c)
                adj.set(r,c, simFunc.sim(v, m.getRowVector(c)));
        }
        return clusterAdjacencyMatrix(adj, method, numClusters);
    }

View Full Code Here

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.