Libsvm3.12版本使用以及二次开发

修改两个类：svm_predict,svm_train,前者是预测的，后者是训练的。

工程图示：

 

根据自己要做实验的数据，修改了一部分返回值的代码：

 

svm_predict.java进行修改为

 

package libsvm.bean;
import libsvm.*;
import java.io.*;
import java.util.*;

public class svm_predict {
	
	private static int  correctCount ;
	private static int  totalCount ;
	
	public  int getTotalCount() {
		return totalCount;
	}

	public int getCorrectCount() {
		return correctCount;
	}
    
	private static double atof(String s)
	{
		return Double.valueOf(s).doubleValue();
	}

	private static int atoi(String s)
	{
		return Integer.parseInt(s);
	}

	private static void predict(BufferedReader input, DataOutputStream output, svm_model model, int predict_probability) throws IOException
	{
		int correct = 0;
		int total = 0;
		double error = 0;
		double sumv = 0, sumy = 0, sumvv = 0, sumyy = 0, sumvy = 0;

		int svm_type=svm.svm_get_svm_type(model);
		int nr_class=svm.svm_get_nr_class(model);
		double[] prob_estimates=null;

		if(predict_probability == 1)
		{
			if(svm_type == svm_parameter.EPSILON_SVR ||
			   svm_type == svm_parameter.NU_SVR)
			{
				System.out.print("Prob. model for test data: target value = predicted value + z,\nz: Laplace distribution e^(-|z|/sigma)/(2sigma),sigma="+svm.svm_get_svr_probability(model)+"\n");
			}
			else
			{
				int[] labels=new int[nr_class];
				svm.svm_get_labels(model,labels);
				prob_estimates = new double[nr_class];
				output.writeBytes("labels");
				for(int j=0;j<nr_class;j++)
					output.writeBytes(" "+labels[j]);
				output.writeBytes("\n");
			}
		}
		while(true)
		{
			String line = input.readLine();
			if(line == null) break;

			StringTokenizer st = new StringTokenizer(line," \t\n\r\f:");

			double target = atof(st.nextToken());
			int m = st.countTokens()/2;
			svm_node[] x = new svm_node[m];
			for(int j=0;j<m;j++)
			{
				x[j] = new svm_node();
				x[j].index = atoi(st.nextToken());
				x[j].value = atof(st.nextToken());
			}

			double v;
			if (predict_probability==1 && (svm_type==svm_parameter.C_SVC || svm_type==svm_parameter.NU_SVC))
			{
				v = svm.svm_predict_probability(model,x,prob_estimates);
				output.writeBytes(v+" ");
				for(int j=0;j<nr_class;j++)
					output.writeBytes(prob_estimates[j]+" ");
				output.writeBytes("\n");
			}
			else //一般进入到这
			{
				v = svm.svm_predict(model,x);
				output.writeBytes(target+"----"+v+"\n");
				/**--如果要取得哪几类分错可以在这里写代码--**/
			}
			
			if(v == target)
			{
				++correct;
	        }
			error += (v-target)*(v-target);
			sumv += v;
			sumy += target;
			sumvv += v*v;
			sumyy += target*target;
			sumvy += v*target;
			++total;
			
		} //end while
		
		if(svm_type == svm_parameter.EPSILON_SVR ||
		   svm_type == svm_parameter.NU_SVR)
		{
			System.out.print("Mean squared error = "+error/total+" (regression)\n");
			System.out.print("Squared correlation coefficient = "+
				 ((total*sumvy-sumv*sumy)*(total*sumvy-sumv*sumy))/
				 ((total*sumvv-sumv*sumv)*(total*sumyy-sumy*sumy))+
				 " (regression)\n");
		}
		else
		{	System.out.print("Accuracy = "+(double)correct/total*100+
				 "% ("+correct+"/"+total+") (classification)\n");
		}
		
		correctCount = correct ;
		totalCount = total ;
	}

	
	private static int predictReturnResult(BufferedReader input, DataOutputStream output, svm_model model, int predict_probability) throws IOException
	{
		int correct = 0;
		int total = 0;
		double error = 0;
		double sumv = 0, sumy = 0, sumvv = 0, sumyy = 0, sumvy = 0;

		int svm_type=svm.svm_get_svm_type(model);
		int nr_class=svm.svm_get_nr_class(model);
		double[] prob_estimates=null;

		if(predict_probability == 1)
		{
			if(svm_type == svm_parameter.EPSILON_SVR ||
			   svm_type == svm_parameter.NU_SVR)
			{
				System.out.print("Prob. model for test data: target value = predicted value + z,\nz: Laplace distribution e^(-|z|/sigma)/(2sigma),sigma="+svm.svm_get_svr_probability(model)+"\n");
			}
			else
			{
				int[] labels=new int[nr_class];
				svm.svm_get_labels(model,labels);
				prob_estimates = new double[nr_class];
				output.writeBytes("labels");
				for(int j=0;j<nr_class;j++)
					output.writeBytes(" "+labels[j]);
				output.writeBytes("\n");
			}
		}
		while(true)
		{
			String line = input.readLine();
			if(line == null) break;

			StringTokenizer st = new StringTokenizer(line," \t\n\r\f:");

			double target = atof(st.nextToken());
			int m = st.countTokens()/2;
			svm_node[] x = new svm_node[m];
			for(int j=0;j<m;j++)
			{
				x[j] = new svm_node();
				x[j].index = atoi(st.nextToken());
				x[j].value = atof(st.nextToken());
			}

			double v;
			if (predict_probability==1 && (svm_type==svm_parameter.C_SVC || svm_type==svm_parameter.NU_SVC))
			{
				v = svm.svm_predict_probability(model,x,prob_estimates);
				output.writeBytes(v+" ");
				for(int j=0;j<nr_class;j++)
					output.writeBytes(prob_estimates[j]+" ");
				output.writeBytes("\n");
			}
			else
			{
				v = svm.svm_predict(model,x);
				output.writeBytes(v+"\n");
			}
			if(v == target)
				++correct;
			error += (v-target)*(v-target);
			sumv += v;
			sumy += target;
			sumvv += v*v;
			sumyy += target*target;
			sumvy += v*target;
			++total;
			
		} //end while
		
		if(svm_type == svm_parameter.EPSILON_SVR ||
		   svm_type == svm_parameter.NU_SVR)
		{
			System.out.print("Mean squared error = "+error/total+" (regression)\n");
			System.out.print("Squared correlation coefficient = "+
				 ((total*sumvy-sumv*sumy)*(total*sumvy-sumv*sumy))/
				 ((total*sumvv-sumv*sumv)*(total*sumyy-sumy*sumy))+
				 " (regression)\n");
		}
		else
		{	System.out.print("Accuracy = "+(double)correct/total*100+
				 "% ("+correct+"/"+total+") (classification)\n");
		}
		
		return correct ;
		
	}
	
	//打印帮助
	private static void exit_with_help()
	{
		System.err.print("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");
		System.exit(1);
	}

	public  void main(String argv[]) throws IOException
	{
		int i, predict_probability=0;

		// parse options
		for(i=0;i<argv.length;i++)
		{
			if(argv[i].charAt(0) != '-') break;
			++i;
			switch(argv[i-1].charAt(1))
			{
				case 'b':
					predict_probability = atoi(argv[i]);
					break;
				default:
					System.err.print("Unknown option: " + argv[i-1] + "\n");
					exit_with_help();
			}
		}
		if(i>=argv.length-2)
			exit_with_help();
		try 
		{
			BufferedReader input = new BufferedReader(new FileReader(argv[i]));
			DataOutputStream output = new DataOutputStream(new BufferedOutputStream(new FileOutputStream(argv[i+2])));
			svm_model model = svm.svm_load_model(argv[i+1]);
			if(predict_probability == 1)
			{
				if(svm.svm_check_probability_model(model)==0)
				{
					System.err.print("Model does not support probabiliy estimates\n");
					System.exit(1);
				}
			}
			else
			{
				if(svm.svm_check_probability_model(model)!=0)
				{
					System.out.print("Model supports probability estimates, but disabled in prediction.\n");
				}
			}
			predict(input,output,model,predict_probability);
			input.close();
			output.close();
		} 
		catch(FileNotFoundException e) 
		{
			exit_with_help();
		}
		catch(ArrayIndexOutOfBoundsException e) 
		{
			exit_with_help();
		}
	}
}

 

 主要增加了返回测试正确与总的数据，如果你想要获取是哪些类分错了，也是可以添加代码的，这里没有标注。

 

svm_train.java 修改后：

 

package libsvm.bean;
import libsvm.*;
import java.io.*;
import java.util.*;

 public class svm_train {
	private svm_parameter param;		// set by parse_command_line
	private svm_problem prob;		// set by read_problem
	private svm_model model;
	private String input_file_name;		// set by parse_command_line
	private String model_file_name;		// set by parse_command_line
	private String error_msg;
	private int cross_validation;
	private int nr_fold;

	private static svm_print_interface svm_print_null = new svm_print_interface()
	{
		public void print(String s) {}
	};

	private static void exit_with_help()
	{
		System.out.print(
		 "Usage: svm_train [options] training_set_file [model_file]\n"
		+"options:\n"
		+"-s svm_type : set type of SVM (default 0)\n"
		+"	0 -- C-SVC\n"
		+"	1 -- nu-SVC\n"
		+"	2 -- one-class SVM\n"
		+"	3 -- epsilon-SVR\n"
		+"	4 -- nu-SVR\n"
		+"-t kernel_type : set type of kernel function (default 2)\n"
		+"	0 -- linear: u'*v\n"
		+"	1 -- polynomial: (gamma*u'*v + coef0)^degree\n"
		+"	2 -- radial basis function: exp(-gamma*|u-v|^2)\n"
		+"	3 -- sigmoid: tanh(gamma*u'*v + coef0)\n"
		+"	4 -- precomputed kernel (kernel values in training_set_file)\n"
		+"-d degree : set degree in kernel function (default 3)\n"
		+"-g gamma : set gamma in kernel function (default 1/num_features)\n"
		+"-r coef0 : set coef0 in kernel function (default 0)\n"
		+"-c cost : set the parameter C of C-SVC, epsilon-SVR, and nu-SVR (default 1)\n"
		+"-n nu : set the parameter nu of nu-SVC, one-class SVM, and nu-SVR (default 0.5)\n"
		+"-p epsilon : set the epsilon in loss function of epsilon-SVR (default 0.1)\n"
		+"-m cachesize : set cache memory size in MB (default 100)\n"
		+"-e epsilon : set tolerance of termination criterion (default 0.001)\n"
		+"-h shrinking : whether to use the shrinking heuristics, 0 or 1 (default 1)\n"
		+"-b probability_estimates : whether to train a SVC or SVR model for probability estimates, 0 or 1 (default 0)\n"
		+"-wi weight : set the parameter C of class i to weight*C, for C-SVC (default 1)\n"
		+"-v n : n-fold cross validation mode\n"
		+"-q : quiet mode (no outputs)\n"
		);
		System.exit(1);
	}

	/**
	 * 做交叉验证
	 */
	private void do_cross_validation()
	{
		int i;
		int total_correct = 0;
		double total_error = 0;
		double sumv = 0, sumy = 0, sumvv = 0, sumyy = 0, sumvy = 0;
		double[] target = new double[prob.l];

		svm.svm_cross_validation(prob,param,nr_fold,target);
		if(param.svm_type == svm_parameter.EPSILON_SVR ||
		   param.svm_type == svm_parameter.NU_SVR)
		{
			for(i=0;i<prob.l;i++)
			{
				double y = prob.y[i];
				double v = target[i];
				total_error += (v-y)*(v-y);
				sumv += v;
				sumy += y;
				sumvv += v*v;
				sumyy += y*y;
				sumvy += v*y;
			}
			System.out.print("Cross Validation Mean squared error = "+total_error/prob.l+"\n");
			System.out.print("Cross Validation Squared correlation coefficient = "+
				((prob.l*sumvy-sumv*sumy)*(prob.l*sumvy-sumv*sumy))/
				((prob.l*sumvv-sumv*sumv)*(prob.l*sumyy-sumy*sumy))+"\n"
				);
		}
		else
		{
			for(i=0;i<prob.l;i++)
				if(target[i] == prob.y[i])
					++total_correct;
			System.out.print("Cross Validation Accuracy = "+100.0*total_correct/prob.l+"%\n");
		}
	}
	
	private void run(String argv[]) throws IOException
	{
		parse_command_line(argv);
		read_problem();
		error_msg = svm.svm_check_parameter(prob,param);

		if(error_msg != null)
		{
			System.err.print("ERROR: "+error_msg+"\n");
			System.exit(1);
		}

		if(cross_validation != 0)
		{
			do_cross_validation();
		}
		else
		{
			model = svm.svm_train(prob,param);
			svm.svm_save_model(model_file_name,model);
		}
	}

	
	/**
	 * 支持向量机训练入口处
	 * @param argv
	 * @throws IOException
	 */
	public  void main(String argv[]) throws IOException
	{
		svm_train t = new svm_train();
		t.run(argv);
	}

	//就是把字符串s转换成double类型 
	private static double atof(String s)
	{
		double d = Double.valueOf(s).doubleValue();
		if (Double.isNaN(d) || Double.isInfinite(d))
		{
			System.err.print("NaN or Infinity in input\n");
			System.exit(1);
		}
		return(d);
	}
	
	//就是把字符串s转换成Int类型 
	private static int atoi(String s)
	{
		return Integer.parseInt(s);
	}
	
   //解析命令行
	private void parse_command_line(String argv[])
	{
		int i;
		svm_print_interface print_func = null;	// default printing to stdout

		param = new svm_parameter();
		
		// default values
		param.svm_type = svm_parameter.C_SVC;
		param.kernel_type = svm_parameter.RBF;
		param.degree = 3;
		param.gamma = 0;	// 1/num_features
		param.coef0 = 0;
		param.nu = 0.5;
		param.cache_size = 100;
		param.C = 1;
		param.eps = 1e-3;
		param.p = 0.1;
		param.shrinking = 1;
		param.probability = 0;
		param.nr_weight = 0;
		param.weight_label = new int[0];
		param.weight = new double[0];
		cross_validation = 0;

		// parse options
		for(i=0;i<argv.length;i++)
		{
			if(argv[i].charAt(0) != '-') break;
			
			if(++i>=argv.length)
				exit_with_help();
			
			switch(argv[i-1].charAt(1))
			{
				case 's':   //支持向量机的类型
					param.svm_type = atoi(argv[i]);
					break;
				case 't':  //支持向量机核函数 
					param.kernel_type = atoi(argv[i]);
					break;
				case 'd':  //支持向量机
					param.degree = atoi(argv[i]);
					break;
				case 'g':  //支持向量机 gamma重要参数,double类型
					param.gamma = atof(argv[i]);
					break;
				case 'r':  //支持向量机 
					param.coef0 = atof(argv[i]);
					break;
				case 'n':
					param.nu = atof(argv[i]);
					break;
				case 'm':
					param.cache_size = atof(argv[i]);
					break;
				case 'c':   //支持向量机C重要参数
					param.C = atof(argv[i]);
					break;
				case 'e':
					param.eps = atof(argv[i]);
					break;
				case 'p':
					param.p = atof(argv[i]);
					break;
				case 'h':
					param.shrinking = atoi(argv[i]);
					break;
				case 'b':
					param.probability = atoi(argv[i]);
					break;
				case 'q':
					print_func = svm_print_null;
					i--;
					break;
				case 'v':
					cross_validation = 1;
					nr_fold = atoi(argv[i]);
					if(nr_fold < 2)
					{
						System.err.print("n-fold cross validation: n must >= 2\n");
						exit_with_help();
					}
					break;
				case 'w': //添加权值
					++param.nr_weight;
					{
						int[] old = param.weight_label;
						param.weight_label = new int[param.nr_weight];
						System.arraycopy(old,0,param.weight_label,0,param.nr_weight-1);
					}

					{
						double[] old = param.weight;
						param.weight = new double[param.nr_weight];
						System.arraycopy(old,0,param.weight,0,param.nr_weight-1);
					}

					param.weight_label[param.nr_weight-1] = atoi(argv[i-1].substring(2));
					param.weight[param.nr_weight-1] = atof(argv[i]);
					break;
				default:
					System.err.print("Unknown option: " + argv[i-1] + "\n");
					exit_with_help();
			} //end switch
		} // end if
		
		svm.svm_set_print_string_function(print_func);
		// determine filenames
		if(i>=argv.length)
			exit_with_help();

		input_file_name = argv[i];

		if(i<argv.length-1)
			model_file_name = argv[i+1];
		else
		{
			int p = argv[i].lastIndexOf('/');
			++p;	// whew...
			model_file_name = argv[i].substring(p)+".model";
		}

	}

	// read in a problem (in svmlight format)
	private void read_problem() throws IOException
	{
		BufferedReader fp = new BufferedReader(new FileReader(input_file_name));
		Vector<Double> vy = new Vector<Double>();
		Vector<svm_node[]> vx = new Vector<svm_node[]>();
		int max_index = 0;

		while(true)
		{
			String line = fp.readLine();
			if(line == null) break;

			StringTokenizer st = new StringTokenizer(line," \t\n\r\f:");

			vy.addElement(atof(st.nextToken()));
			int m = st.countTokens()/2;
			svm_node[] x = new svm_node[m];
			for(int j=0;j<m;j++)
			{
				x[j] = new svm_node();
				x[j].index = atoi(st.nextToken());
				x[j].value = atof(st.nextToken());
			}
			if(m>0) max_index = Math.max(max_index, x[m-1].index);
			vx.addElement(x);
		}

		prob = new svm_problem();
		prob.l = vy.size();
		prob.x = new svm_node[prob.l][];
		for(int i=0;i<prob.l;i++)
			prob.x[i] = vx.elementAt(i);
		prob.y = new double[prob.l];
		for(int i=0;i<prob.l;i++)
			prob.y[i] = vy.elementAt(i);

		if(param.gamma == 0 && max_index > 0)
			param.gamma = 1.0/max_index;

		if(param.kernel_type == svm_parameter.PRECOMPUTED)
			for(int i=0;i<prob.l;i++)
			{
				if (prob.x[i][0].index != 0)
				{
					System.err.print("Wrong kernel matrix: first column must be 0:sample_serial_number\n");
					System.exit(1);
				}
				if ((int)prob.x[i][0].value <= 0 || (int)prob.x[i][0].value > max_index)
				{
					System.err.print("Wrong input format: sample_serial_number out of range\n");
					System.exit(1);
				}
			}

		fp.close();
	}
}

 

也没有什么修改，就是添加了public ，主要修改的还是svm_predict.java这个类。

 

调用的方法：主要的修改参数已经写好了，demo用的测试数据和训练数据是一样的

 

package com.endual.paper.main;

import java.io.IOException;

import libsvm.bean.svm_predict;
import libsvm.bean.svm_train;



/**
 * Libsvm版本3.12 
 * @author endual
 * 导入训练文件，然后训练，导入预测文件，然后预测
 *
 */
public class EndualMain {


	public static  void main(String[] args) throws Exception {
		
		
		/** 建立模型 **/
		String[] arg_train = {"-s","0",               //默认的支持向量机类型
				              "-t","2",               //默认的核函数类型RBF
				              "-d","2",               //set degree in kernel function (default 3)
				              "-g","10",               //set gamma in kernel function (default 1/num_features) 
				              "-r","0",               //set coef0 in kernel function (default 0)
				              "-c","1",               //set the parameter C of C-SVC, epsilon-SVR, and nu-SVR (default 1)
				              "-n","0.5",             //set the parameter nu of nu-SVC, one-class SVM, and nu-SVR (default 0.5)
				              "-b","0",               //whether to train a SVC or SVR model for probability estimates, 0 or 1 (default 0)
				              "file_data\\data\\data.txt",             
				              "file_data\\model\\data_model.txt"} ;
		
		svm_train train = new svm_train();
		train.main(arg_train) ;//进行训练
		
		
		/** 预测数据 **/
		String[] arg_predict = {"-b","0",              //predict probability estimates, 0 or 1 (default 0)
				                "file_data\\data\\data.txt",
				                "file_data\\model\\data_model.txt",
				                "file_data\\out\\data_out.txt"} ;
		
		svm_predict predict = new svm_predict() ;
		predict.main(arg_predict) ;
		int correctCount = predict.getCorrectCount() ; //获取到预测准确的个数
		int totalCount = predict.getTotalCount() ;     //获取到总的预测个数
		System.out.println("||----------------------------------------||");
		System.out.println("|| 预测准确的个数="+correctCount);
		System.out.println("|| 总的预测的个数="+totalCount);
		System.out.println("||--------------------------------------  ||");
		
	}

}

 

运行结果：

..*
optimization finished, #iter = 28
nu = 0.46153846153846156
obj = -4.025964170236124, rho = -0.6825924901648381
nSV = 13, nBSV = 3
..*
optimization finished, #iter = 38
nu = 0.8888888888888888
obj = -8.5069246595277, rho = -0.15000782032015664
nSV = 18, nBSV = 8
..*
optimization finished, #iter = 28
nu = 0.3333333333333333
obj = -2.7925423354383563, rho = -0.7960877909831303
nSV = 12, nBSV = 2
.*
optimization finished, #iter = 15
nu = 0.5454545454545454
obj = -3.9300105118677906, rho = 0.618192908080933
nSV = 11, nBSV = 3
..*
optimization finished, #iter = 13
nu = 0.8
obj = -2.3219227628258747, rho = -0.3342767053776095
nSV = 5, nBSV = 2
.*
optimization finished, #iter = 15
nu = 0.4
obj = -2.7316267112794113, rho = -0.7446069747755875
nSV = 10, nBSV = 2
Total nSV = 23
Accuracy = 100.0% (23/23) (classification)
||----------------------------------------||
|| 预测准确的个数=23
|| 总的预测的个数=23
||--------------------------------------  ||