All/C_CPP/001/svm/svm_model.h

#ifndef __LIBSVM_MODEL_H__
#define __LIBSVM_MODEL_H__

#include <stdint.h>
#define FEATURE_SIZE 200
#define NR_CLASS 2
#define TOTAL_SV 3033
#define NR_CLASSIFER(nc) (nc*(nc-1)/2)

typedef struct {
    uint8_t kernel_type;
    //-- for polynomial kernel -->
    uint8_t degree;
    float gamma;
    float coef0;
    //<---------------------------
    uint8_t nr_class;    /* number of classes, = 2 in regression/one class svm */
    uint32_t total_sv;   /* total #SV */
    float* rhos;          /* constants in decision functions (rho[k*(k-1)/2]) */
    int32_t* label;      /* label of each class (label[k]) */
    float* probA;        /* pariwise probability information */
    float* probB;

    float (*coefs)[TOTAL_SV];  /* coefficients for SVs in decision functions (sv_coef[k-1][l]) */
    uint32_t vec_dim;
    float* SVs;          /* SVs (SV[l]) */
    int32_t* nr_sv;      /* number of SVs for each class (nSV[k]) */
                         /* nSV[0] + nSV[1] + ... + nSV[k-1] = l */
} Model;

/**
 *  init svm params, must be invoked before predict
 */
void init_svm_params(Model *svm_model);

/**
 * two-class only
 */
float predict(const Model *svm_model, const float* x, int len);

/**
 * two-class only
 */
float predict_probability(const Model *svm_model, const float* x, int len);

/**
 * @brief predict multiclasses
 *
 * @param svm_model   SVM model
 * @param x           input x feature
 * @param len         x length
 * @return int32_t    return label
 */
int32_t multi_predict(const Model* svm_model, const float* x, int len);

/**
 * predict multi classes
 * @param[in]  svm_model  SVM model
 * @param[in]  x          input x feature
 * @param[in]  len        x length
 * @param[out] scores_buf probability of each score
 * @param[in]  scores_buf_len  length of scores buf
 */
int32_t multi_predict_probability(const Model *svm_model, const float* x, int len, float* scores_buf, int scores_buf_len);


#endif