AdaBoost_ERP Class Reference

AdaBoost.ERP (AdaBoost.ECC with Re-Partitioning). More...

#include <adaboost_erp.h>

Inheritance diagram for AdaBoost_ERP:

Inheritance graph
[legend]
Collaboration diagram for AdaBoost_ERP:

Collaboration graph
[legend]
List of all members.

Public Member Functions

 AdaBoost_ERP ()
 AdaBoost_ERP (const AdaBoost_ECOC &s)
 AdaBoost_ERP (std::istream &is)
virtual const id_tid () const
virtual AdaBoost_ERPcreate () const
 Create a new object using the default constructor.
virtual AdaBoost_ERPclone () const
 Create a new object by replicating itself.
void set_lr_step (UINT s)

Protected Member Functions

pLearnModel train_with_partial_partition (const ECOC_VECTOR &) const
virtual bool ECOC_partition (UINT, ECOC_VECTOR &) const
virtual pLearnModel train_with_partition (ECOC_VECTOR &) const

Protected Attributes

UINT lrs
 # of steps in the learning / re-partitioning cycle

Detailed Description

AdaBoost.ERP (AdaBoost.ECC with Re-Partitioning).

Alter the ECC table after learning.

Definition at line 19 of file adaboost_erp.h.


Constructor & Destructor Documentation

AdaBoost_ERP  )  [inline]
 

Definition at line 25 of file adaboost_erp.h.

Referenced by AdaBoost_ERP::clone(), and AdaBoost_ERP::create().

AdaBoost_ERP const AdaBoost_ECOC s  )  [inline]
 

Definition at line 26 of file adaboost_erp.h.

AdaBoost_ERP std::istream &  is  )  [inline, explicit]
 

Definition at line 27 of file adaboost_erp.h.


Member Function Documentation

virtual AdaBoost_ERP* clone  )  const [inline, virtual]
 

Create a new object by replicating itself.

Returns:
A pointer to the new copy.
The code for a derived class Derived is always
 return new Derived(*this); 
Though seemingly redundant, it helps to copy an object without knowing the real type of the object.
See also:
C++ FAQ Lite 20.6

Reimplemented from AdaBoost_ECOC.

Definition at line 31 of file adaboost_erp.h.

References AdaBoost_ERP::AdaBoost_ERP().

virtual AdaBoost_ERP* create  )  const [inline, virtual]
 

Create a new object using the default constructor.

The code for a derived class Derived is always

 return new Derived(); 

Reimplemented from AdaBoost_ECOC.

Definition at line 30 of file adaboost_erp.h.

References AdaBoost_ERP::AdaBoost_ERP().

bool ECOC_partition UINT  ,
ECOC_VECTOR
const [protected, virtual]
 

Reimplemented from AdaBoost_ECOC.

Definition at line 12 of file adaboost_erp.cpp.

References AdaBoost_ECOC::ECOC_partition(), MultiClass_ECOC::ECOC_partition(), AdaBoost_ECOC::MAX_2, AdaBoost_ECOC::max_cut_greedy(), AdaBoost_ECOC::par_method, AdaBoost_ECOC::RANDOM_2, and AdaBoost_ECOC::random_half().

virtual const id_t& id  )  const [virtual]
 

Returns:
Class ID string (class name)

Reimplemented from AdaBoost_ECOC.

void set_lr_step UINT  s  )  [inline]
 

Definition at line 34 of file adaboost_erp.h.

References AdaBoost_ERP::lrs.

pLearnModel train_with_partial_partition const ECOC_VECTOR  )  const [protected]
 

Definition at line 33 of file adaboost_erp.cpp.

References dataset::append(), MultiClass_ECOC::ex_class, AdaBoost_ECOC::joint_wgt, Aggregating::lm_base, LearnModel::n_samples, MultiClass_ECOC::nclass, and LearnModel::ptd.

Referenced by AdaBoost_ERP::train_with_partition().

pLearnModel train_with_partition ECOC_VECTOR  )  const [protected, virtual]
 

Reimplemented from AdaBoost_ECOC.

Definition at line 70 of file adaboost_erp.cpp.

References AdaBoost_ECOC::cur_err, MultiClass_ECOC::ex_class, MultiClass_ECOC::is_full_partition(), AdaBoost_ECOC::joint_wgt, AdaBoost_ERP::lrs, LearnModel::n_samples, MultiClass_ECOC::nclass, OUTPUT_PARTITION, LearnModel::ptd, LearnModel::ptw, AdaBoost_ECOC::train_with_full_partition(), and AdaBoost_ERP::train_with_partial_partition().


Member Data Documentation

UINT lrs [protected]
 

# of steps in the learning / re-partitioning cycle

Definition at line 21 of file adaboost_erp.h.

Referenced by AdaBoost_ERP::set_lr_step(), and AdaBoost_ERP::train_with_partition().


The documentation for this class was generated from the following files:
Generated on Wed Nov 8 08:16:37 2006 for LEMGA by  doxygen 1.4.6