用c语言写一个基于svm对葡萄酒品质分类的程序

以下是一个简单的基于SVM的葡萄酒品质分类程序，使用了libsvm库：

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <errno.h>
#include "svm.h"

#define Malloc(type,n) (type *)malloc((n)*sizeof(type))

struct svm_node *x;
int max_nr_attr = 64;

struct svm_model* model;
int predict_probability=0;

void exit_input_error(int line_num)
{
    fprintf(stderr,"Wrong input format at line %d\n", line_num);
    exit(1);
}

void predict(FILE *input, FILE *output)
{
    int correct = 0;
    int total = 0;
    double error = 0;
    double sumv = 0, sumy = 0, sumvv = 0, sumyy = 0, sumvy = 0;

    int svm_type=svm_get_svm_type(model);
    int nr_class=svm_get_nr_class(model);
    double *prob_estimates=NULL;
    int j;

    if (predict_probability)
    {
        if (svm_type==NU_SVR || svm_type==EPSILON_SVR)
            fprintf(output,"labels 1 2 3 actual predicted prob_estimates\n");
        else
            fprintf(output,"labels 1 2 3 ... actual predicted prob_estimates\n");
    }
    else
        fprintf(output,"labels 1 2 3 ... actual predicted\n");

    while(1)
    {
        int i = 0;
        double target_label, predict_label;

        fscanf(input,"%lf",&target_label);
        if(feof(input)) break;

        while(1)
        {
            int index;
            double value;
            fscanf(input,"%d:%lf",&index,&value);
            if(index == -1) break;
            while(index > i)
            {
                x[i].index = i+1;
                x[i].value = 0;
                i++;
            }
            x[i].index = index;
            x[i].value = value;
            i++;
        }
        x[i].index = -1;

        if (predict_probability && (svm_type==C_SVC || svm_type==NU_SVC))
        {
            predict_label = svm_predict_probability(model,x,prob_estimates);
            fprintf(output,"%g ",target_label);
            for(j=0;j<nr_class;j++)
                fprintf(output,"%g ",prob_estimates[j]);
            fprintf(output,"%g\n",predict_label);
        }
        else
        {
            predict_label = svm_predict(model,x);
            fprintf(output,"%g ",target_label);
            fprintf(output,"%g\n",predict_label);
        }

        if(predict_label == target_label)
            ++correct;
        error += (predict_label-target_label)*(predict_label-target_label);
        sumv += predict_label;
        sumy += target_label;
        sumvv += predict_label*predict_label;
        sumyy += target_label*target_label;
        sumvy += predict_label*target_label;
        ++total;
    }
    if (svm_type==NU_SVR || svm_type==EPSILON_SVR)
    {
        printf("Mean squared error = %g (regression)\n",error/total);
        printf("Squared correlation coefficient = %g (regression)\n",
               ((total*sumvy-sumv*sumy)*(total*sumvy-sumv*sumy))/
               ((total*sumvv-sumv*sumv)*(total*sumyy-sumy*sumy))
              );
    }
    else
        printf("Accuracy = %g%% (%d/%d) (classification)\n",
               (double)correct/total*100,correct,total);
    if (predict_probability)
        free(prob_estimates);
}

void exit_with_help()
{
    printf(
    "Usage: svm-predict [options] test_file model_file output_file\n"
    "options:\n"
    "-b probability_estimates: whether to predict probability estimates, 0 or 1 (default 0); one-class SVM not supported yet\n"
    "-q : quiet mode (no outputs)\n"
    );
    exit(1);
}

int main(int argc, char **argv)
{
    char input_file_name[1024];
    char model_file_name[1024];
    char output_file_name[1024];
    FILE *input, *output;

    x = Malloc(struct svm_node,max_nr_attr);

    // parse options
    int i;
    for(i=1;i<argc;i++)
    {
        if(argv[i][0] != '-') break;
        ++i;
        switch(argv[i-1][1])
        {
            case 'b':
                predict_probability = atoi(argv[i]);
                break;
            case 'q':
                svm_print_string = &print_null;
                i--;
                break;
            default:
                fprintf(stderr,"Unknown option: -%c\n", argv[i-1][1]);
                exit_with_help();
        }
    }
    if(i>=argc-2)
        exit_with_help();

    strcpy(input_file_name, argv[i]);
    strcpy(model_file_name, argv[i+1]);
    strcpy(output_file_name, argv[i+2]);

    if((input=fopen(input_file_name,"r"))==NULL)
    {
        fprintf(stderr,"can't open input file %s\n",input_file_name);
        exit(1);
    }

    if((output=fopen(output_file_name,"w"))==NULL)
    {
        fprintf(stderr,"can't open output file %s\n",output_file_name);
        exit(1);
    }

    if((model=svm_load_model(model_file_name))==0)
    {
        fprintf(stderr,"can't open model file %s\n",model_file_name);
        exit(1);
    }

    predict(input,output);
    svm_free_and_destroy_model(&model);
    free(x);
    free(line);
    fclose(input);
    fclose(output);
    return 0;
}

其中，训练数据应该保存为以下格式的文件：

label 1:value 2:value ... -1
label 1:value 2:value ... -1
...

其中，每行表示一个样本，以标签开头，后面跟着特征值。特征值的索引从1开始，以`index:value`的格式表示，最后一个特征值后面必须加上`-1`。

例如，下面是一个样本：

1 1:0.5 2:0.3 3:0.8 -1

表示标签为1，有三个特征，分别为0.5、0.3和0.8。

然后使用以下命令编译程序：

gcc svm-predict.c svm.o -o svm-predict

其中，`svm.o`是libsvm库编译生成的二进制文件，可以从libsvm官网下载到。

然后使用以下命令运行程序：

./svm-predict -b 0 test_file.model model_file output_file

其中，`test_file`是测试数据文件，`model_file`是训练好的模型文件，`output_file`是输出结果文件。 `-b`选项表示是否输出概率估计。