KMeansMiniBatch.h

Go to the documentation of this file.

/*
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 3 of the License, or
 * (at your option) any later version.
 *
 * Written (W) 2014 Parijat Mazumdar
 * Written (W) 2016 Saurabh Mahindre
 */

#ifndef _MBKMEANS_H__
#define _MBKMEANS_H__

#include <shogun/lib/config.h>

#include <shogun/lib/common.h>
#include <shogun/io/SGIO.h>
#include <shogun/distance/Distance.h>
#include <shogun/machine/DistanceMachine.h>
#include <shogun/clustering/KMeansBase.h>

namespace shogun
{
class CKMeansBase;
    
class CKMeansMiniBatch : public CKMeansBase
{
    public:
        CKMeansMiniBatch();

        CKMeansMiniBatch(int32_t k, CDistance* d, bool kmeanspp=false);

        CKMeansMiniBatch(int32_t k_i, CDistance* d_i, SGMatrix<float64_t> centers_i);
        
        virtual ~CKMeansMiniBatch();

        virtual const char* get_name() const { return "KMeansMiniBatch"; }  

        void set_batch_size(int32_t b);

        int32_t get_batch_size() const;

        void set_mb_iter(int32_t t);

        int32_t get_mb_iter() const;

        void set_mb_params(int32_t b, int32_t t);

    protected:

        virtual bool train_machine(CFeatures* data=NULL);

        void minibatch_KMeans();

    private:

        void init_mb_params();

        /* choose b integers between 0 and num-1
         *
         */
        SGVector<int32_t> mbchoose_rand(int32_t b, int32_t num);

    protected:

        int32_t batch_size;

        int32_t minib_iter;

};
}
#endif