edu.ucla.sspace.mains.DVRIWordsiMain 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 2010 Keith Stevens 
 *
 * 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.mains;

import edu.ucla.sspace.common.ArgOptions;
import edu.ucla.sspace.common.SemanticSpaceIO.SSpaceFormat;

import edu.ucla.sspace.dependency.DefaultDependencyPermutationFunction;
import edu.ucla.sspace.dependency.DependencyPermutationFunction;

import edu.ucla.sspace.index.PermutationFunction;
import edu.ucla.sspace.index.RandomIndexVectorGenerator;

import edu.ucla.sspace.util.GeneratorMap;

import edu.ucla.sspace.util.ReflectionUtil;

import edu.ucla.sspace.vector.TernaryVector;

import edu.ucla.sspace.wordsi.DependencyContextGenerator;
import edu.ucla.sspace.wordsi.RandomIndexingDependencyContextGenerator;


/**
 * A dependency based executable class for running {@link Wordsi}.  {@link
 * GenericWordsiMain} provides the core command line arguments and
 * functionality.  This class provides the following additional arguments:
 *
 * 
 *   Optional
 *     
 *        {@code -p}, {@code --pathAcceptor=CLASSNAME} Specifies the {@link
 *       DependencyPathAcceptor} to use when validating paths as features.
 *       (Default: {@link UniversalPathAcceptor})
 *       
 *        {@code -W}, {@code --weightingFunction=CLASSNAME} Specifies the
 *       class that will weight dependency paths.
 *       
 *        {@code -b}, {@code --basisMapping=CLASSNAME} Specifies the class
 *       that deterine what aspect of a {@link DependencyPath} will as a feature
 *       in the word space. (Default: {@link WordBasedBasisMapping})
 *     
 *   
 * 
 *
 * @author Keith Stevens
 */
public class DVRIWordsiMain extends DVWordsiMain {

    /**
     * The {@link DependencyPathBasisMapping} used to generate feature indices
     * for dependency paths.
     */
    private GeneratorMap indexMap;

    /**
     * The {@link DependencyPermutationFunction} repsonsible for permuting index
     * vectors.
     */
    private DependencyPermutationFunction permFunc;

    /**
     * {@inheritDoc}
     */
    protected void addExtraOptions(ArgOptions options) {
        super.addExtraOptions(options);

        options.addOption('P', "permutationFunction",
                          "Specifies the DependencyPermutationFunction for " +
                          "TernaryVectors that will permute index vectors " +
                          "before adding them to context vectors. " +
                          "(Default: None)",
                          true, "CLASSNAME", "Optional");
        options.addOption('l', "vectorLength",
                          "Specifies the length of each index vector. " +
                          "(Default: 5000)",
                          true, "CLASSNAME", "Optional");
    }

    /**
     * {@inheritDoc}
     */
    protected void postProcessing() {
        if (argOptions.hasOption('S')) {
            saveObject(openSaveFile(), indexMap);
            saveObject(openSaveFile(), permFunc);
        }
    }

    /**
     * {@inheritDoc}
     */
    protected DependencyContextGenerator getContextGenerator() {
        int pathLength = argOptions.getIntOption('W', 5);
        int vectorLength = argOptions.getIntOption('l', 5000);

        if (argOptions.hasOption('L')) {
            indexMap = loadObject(openLoadFile());
            permFunc = loadObject(openLoadFile());
        } else {
            indexMap = new GeneratorMap(
                    new RandomIndexVectorGenerator(vectorLength));
            if (argOptions.hasOption('P')) {
                PermutationFunction basePermFunc = 
                    ReflectionUtil.getObjectInstance(
                            argOptions.getStringOption('P'));
                permFunc =
                    new DefaultDependencyPermutationFunction(
                            basePermFunc);
            }
        }

        // Set to read only if in evaluation mode.
        if (argOptions.hasOption('e'))
            throw new Error("fix me");
            //indexMap.setReadOnly(true);

       return new RandomIndexingDependencyContextGenerator(
               permFunc, getAcceptor(), indexMap, vectorLength, pathLength);
    }

    /**
     * {@inheritDoc}
     */
    protected SSpaceFormat getSpaceFormat() {
        return SSpaceFormat.SPARSE_BINARY;
    }
}