edu.ucla.sspace.matrix.YaleSparseMatrix Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of sspace Show documentation
The S-Space Package is a Natural Language Processing library for distributional semantics representations. Distributional semantics representations model the meaning of words, phrases, and sentences as high dimensional vectors or probability distributions. The library includes common algorithms such as Latent Semantic Analysis, Random Indexing, and Latent Dirichlet Allocation. The S-Space package also includes software libraries for matrices, vectors, graphs, and numerous clustering algorithms.
The newest version!
/*
 * Copyright 2009 David Jurgens
 *
 * This file is part of the S-Space package and is covered under the terms and
 * conditions therein.
 *
 * The S-Space package is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as published
 * by the Free Software Foundation and distributed hereunder to you.
 *
 * THIS SOFTWARE IS PROVIDED "AS IS" AND NO REPRESENTATIONS OR WARRANTIES,
 * EXPRESS OR IMPLIED ARE MADE.  BY WAY OF EXAMPLE, BUT NOT LIMITATION, WE MAKE
 * NO REPRESENTATIONS OR WARRANTIES OF MERCHANT- ABILITY OR FITNESS FOR ANY
 * PARTICULAR PURPOSE OR THAT THE USE OF THE LICENSED SOFTWARE OR DOCUMENTATION
 * WILL NOT INFRINGE ANY THIRD PARTY PATENTS, COPYRIGHTS, TRADEMARKS OR OTHER
 * RIGHTS.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program. If not, see .
 */

package edu.ucla.sspace.matrix;

import edu.ucla.sspace.vector.CompactSparseVector;
import edu.ucla.sspace.vector.DoubleVector;
import edu.ucla.sspace.vector.SparseDoubleVector;
import edu.ucla.sspace.vector.SparseHashDoubleVector;
import edu.ucla.sspace.vector.Vectors;


/**
 * A sparse {@code Matrix} based on the Yale Sparse Matrix Format, as
 * implemented in {@link CompactSparseVector}.  Each row is allocated a pair of
 * arrays which keeps the non-zero column values in column order.  Lookups are
 * O(log n) where n is the number of non-zero values for the largest row.  The
 * size of this matrix is fixed, and attempts to access rows or columns beyond
 * the size will throw an {@link IndexOutOfBoundsException}.
 *
 * @author David Jurgens
 */
public class YaleSparseMatrix implements SparseMatrix, java.io.Serializable {

    private static final long serialVersionUID = 1L;

    /**
     * The number of rows contained in this {@code SparseMatrix}.
     */
    private final int rows;

    /**
     * The number of columns contained in this {@code SparseMatrix}.
     */
    private final int cols;

    /**
     * Each row is defined as a {@link CompactSparseVector} which does most of
     * the work.
     */
    private final CompactSparseVector[] sparseMatrix;

    /**
     * Constructs a sparse matrix with the specified dimensions.
     */
    public YaleSparseMatrix(int rows, int cols) {
        this.rows = rows;
        this.cols = cols;
        sparseMatrix = new CompactSparseVector[rows];
        for (int i = 0; i < rows; ++i)
            sparseMatrix[i] = new CompactSparseVector(cols);
    }

    /**
     * Constructs a sparse matrix using values from the given two dimension
     * array.  Only non-zero values will be stored.
     *
     * @throws IllegalArgumentException If either the number of rows is equal to
     *         0, or the column lengths are jagged
     */
    public YaleSparseMatrix(double[][] values) {
        if (values.length == 0)
            throw new IllegalArgumentException(
                    "Matrix must have non zero size");

        this.rows = values.length;
        this.cols = values[0].length;

        sparseMatrix = new CompactSparseVector[rows];
        for (int r = 0; r < rows; ++r) {
            if (values[r].length != cols)
                throw new IllegalArgumentException(
                        "Cannot form matrix from jagged array");

            sparseMatrix[r] = new CompactSparseVector(cols);
            for (int c = 0; c < cols; ++c)
                if (values[r][c] != 0d)
                    set(r, c, values[r][c]);
        }
    }

    /**
     * Checks that the indices are within the bounds of this matrix or throws an
     * {@link IndexOutOfBoundsException} if not.
     */        
    private void checkIndices(int row, int col) {
        if (row < 0 || col < 0 || row >= rows || col >= cols) {
            throw new IndexOutOfBoundsException();
        }
    }

    /**
     * {@inheritDoc}
     */
    public double get(int row, int col) {
        checkIndices(row, col);
        return sparseMatrix[row].get(col);
    }

    /**
     * {@inheritDoc}
     */
    public double[] getColumn(int column) {
        double[] values = new double[rows];
        for (int row = 0; row < rows; ++row)
            values[row] = get(row, column);
        return values;
    }

    /**
     * {@inheritDoc}
     */
    public SparseDoubleVector getColumnVector(int column) {
        SparseDoubleVector values = new SparseHashDoubleVector(rows);
        for (int row = 0; row < rows; ++row)
            values.set(row, get(row, column));
        return values;
    }

    /**
     * {@inheritDoc}
     */
    public double[] getRow(int row) {
        return sparseMatrix[row].toArray();
    }

    /**
     * {@inheritDoc}
     */
    public SparseDoubleVector getRowVector(int row) {
        return Vectors.immutable(sparseMatrix[row]);
    }

    /**
     * {@inheritDoc}
     */
    public int columns() {
        return cols;
    }

    /**
     * {@inheritDoc}
     */
    public void set(int row, int col, double val) {
        checkIndices(row, col);
        sparseMatrix[row].set(col, val);
    }

    /**
     * {@inheritDoc}
     */
    public void setColumn(int column, double[] values) {
        if (values.length != rows) {
            throw new IllegalArgumentException(
                    "invalid number of rows: " + values.length);
        }
        for (int row = 0; row < rows; ++row)
            set(row, column, values[row]);
    }

    /**
     * {@inheritDoc}
     */
    public void setColumn(int column, DoubleVector values) {
        if (values.length() != rows) {
            throw new IllegalArgumentException(
                    "invalid number of rows: " + values.length());
        }
        for (int row = 0; row < rows; ++row)
            set(row, column, values.get(row));
    }

    /**
     * {@inheritDoc}
     */
    public void setRow(int row, double[] columns) {
        if (columns.length != cols) {
            throw new IllegalArgumentException(
                "invalid number of columns: " + columns.length);
        }
        for (int col = 0; col < cols; ++col) {
            sparseMatrix[row].set(col, columns[col]);
        }
    }

    /**
     * {@inheritDoc}
     */
    public void setRow(int row, DoubleVector values) {
        if (values.length() != cols) {
            throw new IllegalArgumentException(
                "invalid number of columns: " + values.length());
        }
        Vectors.copy(sparseMatrix[row], values);
    }

    /**
     * {@inheritDoc}
     */
    public double[][] toDenseArray() {
        double[][] m = new double[rows][cols];
        for (int r = 0; r < rows; ++r) {
            m[r] = sparseMatrix[r].toArray();
        }
        return m;
    }

    /**
     * {@inheritDoc}
     */
    public int rows() {
        return rows;
    }
}