One_Class_SVM::~One_Class_SVM(){
if ( trained )
{
svm_free_model_content( _model );
free( _model );
}
else
{
SVM_Module::free_support_vectors();
}
delete _class_labels;
delete _support_vectors;
delete _parameters;
}
void fillAge(list_t *passengerList,list_t *testPassengerList){
list_t inAge,notAge;
struct svm_parameter param; // set by parse_command_line
struct svm_problem prob,prob2; // set by read_problem
struct svm_model *model;
struct svm_node *x_space;
int max_index=5;
int i;
cell_t *cur;
passenger *human;
// default values
param.svm_type = NU_SVR;
param.kernel_type = LINEAR;
param.degree = 3;
param.gamma = 0; // 1/num_features
param.coef0 = 0;
param.nu = 0.5;
param.cache_size = 0;
param.C = 1;
param.eps = 1e-2;
param.p = 0.1;
param.shrinking = 1;
param.probability = 0;
param.nr_weight = 0;
param.weight_label = NULL;
param.weight = NULL;
memset(&inAge,0,sizeof(list_t));
memset(¬Age,0,sizeof(list_t));
for(cur=passengerList->first;cur;cur=cur->next){
human=(passenger *)cur->data;
if(human->age==-1) addList(¬Age,human);
else addList(&inAge,human);
}
for(cur=testPassengerList->first;cur;cur=cur->next){
human=(passenger *)cur->data;
if(human->age==-1) addList(¬Age,human);
else addList(&inAge,human);
}
prob.l=inAge.size;
prob.y = Malloc(double,prob.l);
prob.x = Malloc(struct svm_node *,prob.l);
for(i=0,cur=inAge.first;cur;cur=cur->next,i++){
x_space = Malloc(struct svm_node,max_index+1);
passenger *human=(passenger *)cur->data;
x_space[0].index=1;
x_space[0].value=(double)human->sex;
x_space[1].index=2;
x_space[1].value=(double)human->rank;
x_space[2].index=3;
x_space[2].value=(double)human->fare;
x_space[3].index=4;
x_space[3].value=(double)human->prop1;
x_space[4].index=5;
x_space[4].value=getNameNo(human->name->honorific);
x_space[max_index].index = -1;
prob.x[i]=x_space;
prob.y[i]=(double)human->age;
}
prob2.l=notAge.size;
prob2.x = Malloc(struct svm_node *,prob2.l);
passenger **tmp= Malloc(passenger *,prob2.l);
for(i=0,cur=notAge.first;cur;cur=cur->next,i++){
x_space = Malloc(struct svm_node,max_index+1);
passenger *human=(passenger *)cur->data;
x_space[0].index=1;
x_space[0].value=(double)human->sex;
x_space[1].index=2;
x_space[1].value=(double)human->rank;
x_space[2].index=3;
x_space[2].value=(double)human->fare;
x_space[3].index=4;
x_space[3].value=(double)human->prop1;
x_space[4].index=5;
x_space[4].value=getNameNo(human->name->honorific);
x_space[max_index].index = -1;
prob2.x[i]=x_space;
tmp[i]=human;
}
param.gamma=1.0/(double)max_index;
scale(prob.x,prob.l,prob2.x,prob2.l,max_index);
//printf("%lf\n",cv(¶m,&prob));exit(0);
model = svm_train(&prob,¶m);
for(i=0;i<prob2.l;i++){
tmp[i]->age=(int)svm_predict(model,prob2.x[i]);
}
svm_free_model_content(model);
svm_free_and_destroy_model(&model);
svm_destroy_param(¶m);
for(i=0,cur=inAge.first;cur;cur=cur->next,i++){
free(prob.x[i]);
}
for(i=0,cur=notAge.first;cur;cur=cur->next,i++){
free(prob2.x[i]);
}
free(prob.x);
free(prob.y);
free(prob2.x);
}
/// <summary>
/// use a kernel matrix and solve the svm problem
/// </summary>
/// <param name="Kernel">precomputed kernel</param>
DualSolution libSVMWrapper::Solve(const Eigen::MatrixXd &Kernel)
{
// unfortunately libsvm needs "svm_nodes", so we have to copy everything
// workaround pointer to matrix entries (eigen3 does not allow it?)
// copy entries (code from libSVM)
int j = 0, sc = NumberOfData + 1;
// TODO
// libSVM_x_space[j+k].value = *(Kernel.data() + (k-1)*NumberOfData + i); // by reference or pointer
#pragma omp parallel for
for (int i = 0; i < libSVM_Problem.l; i++)
{
j = (sc+1)*i ;
for (int k = 0; k < sc; k++)
{
libSVM_x_space[j].index = k + 1;
if (k == 0)
{
libSVM_x_space[j].value = i + 1;
}
else
{
libSVM_x_space[j+k].value = Kernel(i, k - 1);
}
}
j = ((sc+1)*i+sc) ;
libSVM_x_space[j+1].index = -1;
}
#ifdef DEBUG
for (int i = 0; i < libSVM_Problem.l; i++)
{
if ((int) libSVM_Problem.x[i][0].value <= 0 || (int) libSVM_Problem.x[i][0].value > sc)
{
printf("Wrong input format: sample_serial_number out of range\n");
exit(0);
}
}
const char *error_msg;
error_msg = svm_check_parameter(&libSVM_Problem, &libSVM_Parameter);
if (error_msg)
{
fprintf(stderr, "ERROR: %s\n", error_msg);
exit(1);
}
#endif
// train the model
if (libSVM_Model != NULL)
{
svm_free_model_content(libSVM_Model);
libSVM_Model = NULL;
}
libSVM_Model = svm_train(&libSVM_Problem, &libSVM_Parameter);
// extract results
// bias is easy
double Bias = -1 * libSVM_Model->rho[0];
// alpha should be dense now
Eigen::VectorXd Alpha = Eigen::MatrixXd::Zero(NumberOfData, 1);
for (int i = 0; i < libSVM_Model->l; i++)
{
Alpha(libSVM_Model->sv_indices[i] - 1) = (libSVM_Model->sv_coef[0][i] < 0) ? -1 * libSVM_Model->sv_coef[0][i] : libSVM_Model->sv_coef[0][i];
}
DualSolution DS;
DS.Bias = Bias;
DS.Alpha = Alpha;
Eigen::VectorXd tt = Alpha.cwiseProduct(Y);
// objective value of dual solution
DS.Value = Alpha.sum() - 0.5* (double)(tt.transpose() * (Kernel*tt));
return DS;
}
