Public Member Functions | Protected Member Functions | Protected Attributes

CLPBoost Class Reference

Detailed Description

Class LPBoost trains a linear classifier called Linear Programming Machine, i.e. a SVM using a $\ell_1$ norm regularizer.

It solves the following optimization problem using Boosting on the input features:

\begin{eqnarray*} \min_{{\bf w}={(\bf w^+},{\bf w^-}), b, {\bf \xi}} && \sum_{i=1}^N ( {\bf w}^+_i + {\bf w}^-_i) + C \sum_{i=1}^{N} \xi_i\\ \mbox{s.t.} && -y_i(({\bf w}^+-{\bf w}^-)^T {\bf x}_i + b)-{\bf \xi}_i \leq -1\\ && \quad {\bf x}_i \geq 0\\\ && {\bf w}_i \geq 0,\quad \forall i=1\dots N \end{eqnarray*}

Note that currently CPLEX is required to solve this problem. This implementation is faster than solving the linear program directly in CPLEX (as was done in CLPM).

See also:

Definition at line 48 of file LPBoost.h.

List of all members.

Public Member Functions

 CLPBoost ()
virtual ~CLPBoost ()
virtual EClassifierType get_classifier_type ()
bool init (int32_t num_vec)
void cleanup ()
virtual void set_features (CDotFeatures *feat)
void set_C (float64_t c_neg, float64_t c_pos)
float64_t get_C1 ()
float64_t get_C2 ()
void set_bias_enabled (bool enable_bias)
bool get_bias_enabled ()
void set_epsilon (float64_t eps)
float64_t get_epsilon ()
float64_t find_max_violator (int32_t &max_dim)
virtual const char * get_name () const

Protected Member Functions

virtual bool train_machine (CFeatures *data=NULL)

Protected Attributes

float64_t C1
float64_t C2
bool use_bias
float64_t epsilon
CDynamicArray< int32_t > * dim
int32_t num_sfeat
int32_t num_svec
SGSparseVector< float64_t > * sfeat

Constructor & Destructor Documentation

CLPBoost (  ) 

Definition at line 25 of file LPBoost.cpp.

~CLPBoost (  )  [virtual]

Definition at line 36 of file LPBoost.cpp.

Member Function Documentation

void cleanup (  ) 

Definition at line 57 of file LPBoost.cpp.

float64_t find_max_violator ( int32_t &  max_dim  ) 

Definition at line 69 of file LPBoost.cpp.

bool get_bias_enabled (  ) 

Definition at line 87 of file LPBoost.h.

float64_t get_C1 (  ) 

Definition at line 83 of file LPBoost.h.

float64_t get_C2 (  ) 

Definition at line 84 of file LPBoost.h.

virtual EClassifierType get_classifier_type (  )  [virtual]

Definition at line 54 of file LPBoost.h.

float64_t get_epsilon (  ) 

Definition at line 90 of file LPBoost.h.

virtual const char* get_name (  )  const [virtual]
object name

Definition at line 95 of file LPBoost.h.

bool init ( int32_t  num_vec  ) 

Definition at line 41 of file LPBoost.cpp.

void set_bias_enabled ( bool  enable_bias  ) 

Definition at line 86 of file LPBoost.h.

void set_C ( float64_t  c_neg,
float64_t  c_pos 

set C

c_neg new C constant for negatively labeled examples
c_pos new C constant for positively labeled examples

Definition at line 81 of file LPBoost.h.

void set_epsilon ( float64_t  eps  ) 

Definition at line 89 of file LPBoost.h.

virtual void set_features ( CDotFeatures feat  )  [virtual]

set features

feat features to set

Definition at line 66 of file LPBoost.h.

bool train_machine ( CFeatures data = NULL  )  [protected, virtual]

train classifier

data training data (parameter can be avoided if distance or kernel-based classifiers are used and distance/kernels are initialized with train data)
whether training was successful

Definition at line 104 of file LPBoost.cpp.

Member Data Documentation

float64_t C1 [protected]

Definition at line 109 of file LPBoost.h.

float64_t C2 [protected]

Definition at line 110 of file LPBoost.h.

CDynamicArray<int32_t>* dim [protected]

Definition at line 115 of file LPBoost.h.

float64_t epsilon [protected]

Definition at line 112 of file LPBoost.h.

int32_t num_sfeat [protected]

Definition at line 117 of file LPBoost.h.

int32_t num_svec [protected]

Definition at line 118 of file LPBoost.h.

Definition at line 119 of file LPBoost.h.

float64_t* u [protected]

Definition at line 114 of file LPBoost.h.

bool use_bias [protected]

Definition at line 111 of file LPBoost.h.

The documentation for this class was generated from the following files:
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines

SHOGUN Machine Learning Toolbox - Documentation