edu.ucla.sspace.util.SparseIntHashArray Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of sspace-wordsi Show documentation

The S-Space Package is a collection of algorithms for building Semantic Spaces as well as a highly-scalable library for designing new distributional semantics algorithms. Distributional algorithms process text corpora and represent the semantic for words as high dimensional feature vectors. This package also includes matrices, vectors, and numerous clustering algorithms. These approaches are known by many names, such as word spaces, semantic spaces, or distributed semantics and rest upon the Distributional Hypothesis: words that appear in similar contexts have similar meanings.

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.util;

import java.io.Serializable;

import java.util.HashMap;
import java.util.Map;


/**
 * A sparse {@code int} array.  This class offers amortized constant time
 * access to array indices at the expense of space.
 *
 * Instance offer a space savings of retaining only the non-zero indices and
 * values.  For large array with only a few values set, this offers a huge
 * savings.  However, as the cardinality of the array grows in relation to its
 * size, a dense {@code int[]} array will offer better performance in both space
 * and time.  This is especially true if the sparse array instance approaches a
 * cardinality to size ratio of {@code .5}.
 *
 * This class offers much better performance than {@link SparseIntArray}, but
 * will use significantly more space as the cardinality increases.  In addition,
 * this class will marshall {@code int} primitives into {@code Integer} objects.
 *
 * @see SparseArray
 * @see SparseIntArray
 */
public class SparseIntHashArray implements SparseArray, Serializable {

    private static final long serialVersionUID = 1L;

    /**
     * The maximum length of this array
     */
    private final int maxLength;
  
    /**
     * A mapping from an array index to its value
     */
    private Map indexToValue;

    /**
     * Creates a sparse {@code int} array that grows to the maximum size set by
     * {@link Integer#MAX_VALUE}.
     */
    public SparseIntHashArray() {
	this(Integer.MAX_VALUE);
    }

    /**
     * Creates a sparse {@code int} array with a fixed length
     */
    public SparseIntHashArray(int length) {
	if (length < 0)
	    throw new IllegalArgumentException("length must be non-negative");
	maxLength = length;
	
        indexToValue = new HashMap();
    }

    /**
     * Creates a sparse array copy of the provided array, retaining only the
     * non-zero values.  The length of the provided array is used to set the
     * maximum size of this sparse array.
     */
    public SparseIntHashArray(int[] array) {

	maxLength = array.length;

	// Find how many non-zero elements there are
	int nonZero = 0;
	for (int i = 0; i < array.length; ++i) {
	    if (array[i] != 0)
		indexToValue.put(i, array[i]);
	}
    }

    /**
     * {@inheritDoc}
     */
    public int cardinality() {
	return indexToValue.size();
    }

    /**
     * {@inheritDoc}
     */ 
    public Integer get(int index) {
	Integer i = indexToValue.get(index);
        return (i == null) ? 0 : i;
    }

    /**
     * Returns the indices of the array that contain non-{@code 0} values.
     *
     * @return the indices that contain values
     */
    public int[] getElementIndices() {
        Integer[] indices = indexToValue.keySet().toArray(new Integer[0]);
        int[] primitive = new int[indices.length];
        for (int i = 0; i < indices.length; ++i)
            primitive[i] = indices[i].intValue();
        return primitive;
    }
    
    /**
     * Returns the maximum length of this array.
     */
    public int length() {
	return maxLength;
    }

    /**
     * {@inheritDoc}
     */
    public void set(int index, Integer value) {
        if (value == 0)
            indexToValue.remove(index);
        else
            indexToValue.put(index, value);
    }

    /**
     * {@inheritDoc}
     */
    @SuppressWarnings("unchecked")
    public  E[] toArray(E[] array) {
	for (int i = 0; i < array.length; ++i) {
            Integer j = indexToValue.get(i);
            if (j != null)
                array[i] = (E)j;
	}
	return array;
    }
}