fann_train_data.c

python 神经网络 数据挖掘 python实现的神经网络算法

/*
 * Fast Artificial Neural Network Library (fann) Copyright (C) 2003
 * Steffen Nissen ([e-mail])
 * 
 * This library is free software; you can redistribute it and/or modify it 
 * under the terms of the GNU Lesser General Public License as published
 * by the Free Software Foundation; either version 2.1 of the License, or
 * (at your option) any later version.
 * 
 * This library is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 * 
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 
 */

#include <stdio.h>
#include <stdlib.h>
#include <stdarg.h>
#include <string.h>

#include "config.h"
#include "fann.h"

/*
 * Reads training data from a file. 
 */
FANN_EXTERNAL struct fann_train_data *FANN_API fann_read_train_from_file(const char *configuration_file)
{
	struct fann_train_data *data;
	FILE *file = fopen(configuration_file, "r");

	if(!file)
	{
		fann_error(NULL, FANN_E_CANT_OPEN_CONFIG_R, configuration_file);
		return NULL;
	}

	data = fann_read_train_from_fd(file, configuration_file);
	fclose(file);
	return data;
}

/*
 * Save training data to a file 
 */
FANN_EXTERNAL int FANN_API fann_save_train(struct fann_train_data *data, const char *filename)
{
	return fann_save_train_internal(data, filename, 0, 0);
}

/*
 * Save training data to a file in fixed point algebra. (Good for testing
 * a network in fixed point) 
 */
FANN_EXTERNAL int FANN_API fann_save_train_to_fixed(struct fann_train_data *data, const char *filename,
													 unsigned int decimal_point)
{
	return fann_save_train_internal(data, filename, 1, decimal_point);
}

/*
 * deallocate the train data structure. 
 */
FANN_EXTERNAL void FANN_API fann_destroy_train(struct fann_train_data *data)
{
	if(data == NULL)
		return;
	if(data->input != NULL)
		fann_safe_free(data->input[0]);
	if(data->output != NULL)
		fann_safe_free(data->output[0]);
	fann_safe_free(data->input);
	fann_safe_free(data->output);
	fann_safe_free(data);
}

/*
 * Test a set of training data and calculate the MSE 
 */
FANN_EXTERNAL float FANN_API fann_test_data(struct fann *ann, struct fann_train_data *data)
{
	unsigned int i;

	fann_reset_MSE(ann);

	for(i = 0; i != data->num_data; i++)
	{
		fann_test(ann, data->input[i], data->output[i]);
	}

	return fann_get_MSE(ann);
}

/*
 * Creates training data from a callback function.
 */
FANN_EXTERNAL struct fann_train_data * FANN_API fann_create_train_from_callback(unsigned int num_data,
                                          unsigned int num_input,
                                          unsigned int num_output,
                                          void (FANN_API *user_function)( unsigned int,
                                                                 unsigned int,
                                                                 unsigned int,
                                                                 fann_type * ,
                                                                 fann_type * ))
{
    unsigned int i;
    fann_type *data_input, *data_output;
    struct fann_train_data *data = (struct fann_train_data *)
                                                                                 malloc(sizeof(struct fann_train_data));

    if(data == NULL){
        fann_error(NULL, FANN_E_CANT_ALLOCATE_MEM);
        return NULL;
    }

    fann_init_error_data((struct fann_error *) data);

    data->num_data     = num_data;
    data->num_input    = num_input;
    data->num_output = num_output;

    data->input = (fann_type **) calloc(num_data, sizeof(fann_type *));
    if(data->input == NULL)
    {
        fann_error(NULL, FANN_E_CANT_ALLOCATE_MEM);
        fann_destroy_train(data);
        return NULL;
    }

    data->output = (fann_type **) calloc(num_data, sizeof(fann_type *));
    if(data->output == NULL)
    {
        fann_error(NULL, FANN_E_CANT_ALLOCATE_MEM);
        fann_destroy_train(data);
        return NULL;
    }

    data_input = (fann_type *) calloc(num_input * num_data, sizeof(fann_type));
    if(data_input == NULL)
    {
        fann_error(NULL, FANN_E_CANT_ALLOCATE_MEM);
        fann_destroy_train(data);
        return NULL;
    }

    data_output = (fann_type *) calloc(num_output * num_data, sizeof(fann_type));
    if(data_output == NULL)
    {
        fann_error(NULL, FANN_E_CANT_ALLOCATE_MEM);
        fann_destroy_train(data);
        return NULL;
    }

    for( i = 0; i != num_data; i++)
    {
        data->input[i] = data_input;
        data_input += num_input;

        data->output[i] = data_output;
        data_output += num_output;

        (*user_function)(i, num_input, num_output, data->input[i],data->output[i] );
    }

    return data;
} 

#ifndef FIXEDFANN

/*
 * Internal train function 
 */
float fann_train_epoch_quickprop(struct fann *ann, struct fann_train_data *data)
{
	unsigned int i;

	if(ann->prev_train_slopes == NULL)
	{
		fann_clear_train_arrays(ann);
	}

	fann_reset_MSE(ann);

	for(i = 0; i < data->num_data; i++)
	{
		fann_run(ann, data->input[i]);
		fann_compute_MSE(ann, data->output[i]);
		fann_backpropagate_MSE(ann);
		fann_update_slopes_batch(ann, ann->first_layer + 1, ann->last_layer - 1);
	}
	fann_update_weights_quickprop(ann, data->num_data, 0, ann->total_connections);

	return fann_get_MSE(ann);
}

/*
 * Internal train function 
 */
float fann_train_epoch_irpropm(struct fann *ann, struct fann_train_data *data)
{
	unsigned int i;

	if(ann->prev_train_slopes == NULL)
	{
		fann_clear_train_arrays(ann);
	}

	fann_reset_MSE(ann);

	for(i = 0; i < data->num_data; i++)
	{
		fann_run(ann, data->input[i]);
		fann_compute_MSE(ann, data->output[i]);
		fann_backpropagate_MSE(ann);
		fann_update_slopes_batch(ann, ann->first_layer + 1, ann->last_layer - 1);
	}

	fann_update_weights_irpropm(ann, 0, ann->total_connections);

	return fann_get_MSE(ann);
}

/*
 * Internal train function 
 */
float fann_train_epoch_batch(struct fann *ann, struct fann_train_data *data)
{
	unsigned int i;

	fann_reset_MSE(ann);

	for(i = 0; i < data->num_data; i++)
	{
		fann_run(ann, data->input[i]);
		fann_compute_MSE(ann, data->output[i]);
		fann_backpropagate_MSE(ann);
		fann_update_slopes_batch(ann, ann->first_layer + 1, ann->last_layer - 1);
	}

	fann_update_weights_batch(ann, data->num_data, 0, ann->total_connections);

	return fann_get_MSE(ann);
}

/*
 * Internal train function 
 */
float fann_train_epoch_incremental(struct fann *ann, struct fann_train_data *data)
{
	unsigned int i;

	fann_reset_MSE(ann);

	for(i = 0; i != data->num_data; i++)
	{
		fann_train(ann, data->input[i], data->output[i]);
	}

	return fann_get_MSE(ann);
}

/*
 * Train for one epoch with the selected training algorithm 
 */
FANN_EXTERNAL float FANN_API fann_train_epoch(struct fann *ann, struct fann_train_data *data)
{
	switch (ann->training_algorithm)
	{
	case FANN_TRAIN_QUICKPROP:
		return fann_train_epoch_quickprop(ann, data);
	case FANN_TRAIN_RPROP:
		return fann_train_epoch_irpropm(ann, data);
	case FANN_TRAIN_BATCH:
		return fann_train_epoch_batch(ann, data);
	case FANN_TRAIN_INCREMENTAL:
		return fann_train_epoch_incremental(ann, data);
	}
	return 0;
}

FANN_EXTERNAL void FANN_API fann_train_on_data(struct fann *ann, struct fann_train_data *data,
											   unsigned int max_epochs,
											   unsigned int epochs_between_reports,
											   float desired_error)
{
	float error;
	unsigned int i;
	int desired_error_reached;

#ifdef DEBUG
	printf("Training with %s\n", FANN_TRAIN_NAMES[ann->training_algorithm]);
#endif

	if(epochs_between_reports && ann->callback == NULL)
	{
		printf("Max epochs %8d. Desired error: %.10f.\n", max_epochs, desired_error);
	}

	for(i = 1; i <= max_epochs; i++)
	{
		/*
		 * train 
		 */
		error = fann_train_epoch(ann, data);
		desired_error_reached = fann_desired_error_reached(ann, desired_error);

		/*
		 * print current output 
		 */
		if(epochs_between_reports &&
		   (i % epochs_between_reports == 0 || i == max_epochs || i == 1 ||
			desired_error_reached == 0))
		{
			if(ann->callback == NULL)
			{
				printf("Epochs     %8d. Current error: %.10f. Bit fail %d.\n", i, error,
					   ann->num_bit_fail);
			}
			else if(((*ann->callback)(ann, data, max_epochs, epochs_between_reports, 
									  desired_error, i)) == -1)
			{
				/*
				 * you can break the training by returning -1 
				 */
				break;
			}
		}

		if(desired_error_reached == 0)
			break;
	}
}

FANN_EXTERNAL void FANN_API fann_train_on_file(struct fann *ann, const char *filename,
											   unsigned int max_epochs,
											   unsigned int epochs_between_reports,
											   float desired_error)
{
	struct fann_train_data *data = fann_read_train_from_file(filename);

	if(data == NULL)
	{
		return;
	}
	fann_train_on_data(ann, data, max_epochs, epochs_between_reports, desired_error);
	fann_destroy_train(data);
}

#endif

/*
 * shuffles training data, randomizing the order 
 */
FANN_EXTERNAL void FANN_API fann_shuffle_train_data(struct fann_train_data *train_data)
{
	unsigned int dat = 0, elem, swap;
	fann_type temp;

	for(; dat < train_data->num_data; dat++)
	{
		swap = (unsigned int) (rand() % train_data->num_data);
		if(swap != dat)
		{
			for(elem = 0; elem < train_data->num_input; elem++)
			{
				temp = train_data->input[dat][elem];
				train_data->input[dat][elem] = train_data->input[swap][elem];
				train_data->input[swap][elem] = temp;
			}
			for(elem = 0; elem < train_data->num_output; elem++)
			{
				temp = train_data->output[dat][elem];
				train_data->output[dat][elem] = train_data->output[swap][elem];
				train_data->output[swap][elem] = temp;
			}
		}
	}
}

/*
 * INTERNAL FUNCTION Scales data to a specific range 
 */
void fann_scale_data(fann_type ** data, unsigned int num_data, unsigned int num_elem,
					 fann_type new_min, fann_type new_max)
{
	unsigned int dat, elem;
	fann_type old_min, old_max, temp, old_span, new_span, factor;

	old_min = old_max = data[0][0];

	/*
	 * first calculate min and max 
	 */
	for(dat = 0; dat < num_data; dat++)
	{
		for(elem = 0; elem < num_elem; elem++)
		{
			temp = data[dat][elem];
			if(temp < old_min)
				old_min = temp;
			else if(temp > old_max)
				old_max = temp;
		}
	}

	old_span = old_max - old_min;
	new_span = new_max - new_min;