theplu::yat::classifier::BootstrapSampler Class Reference

Class creating trainingset and validationset using bootstrapping. More...

#include <yat/classifier/BootstrapSampler.h>

Inheritance diagram for theplu::yat::classifier::BootstrapSampler:

theplu::yat::classifier::Sampler

List of all members.

Public Member Functions

 BootstrapSampler (const Target &target, const size_t N)
 Constructor.
virtual ~BootstrapSampler ()
size_t size (void) const
const Targettarget (void) const
const utility::Indextraining_index (size_t i) const
const Targettraining_target (size_t i) const
const utility::Indexvalidation_index (size_t i) const
const Targetvalidation_target (size_t i) const

Protected Attributes

std::vector< utility::Indextraining_index_
 index of training sets for the partitions
std::vector< Targettraining_target_
 Targets for training sets for the partitions.
std::vector< utility::Indexvalidation_index_
 index of validation sets for the partitions
std::vector< Targetvalidation_target_
 Targets for validation sets for the partitions.


Detailed Description

Class creating trainingset and validationset using bootstrapping.

This is done in a balanced way, meaning the proportions between the classes in the trainingset is equal to the proportions in the whole dataset. For each class $ N_i $ samples are drawn with replacement, where $ N_i $ is number of samples in class i.


Constructor & Destructor Documentation

theplu::yat::classifier::BootstrapSampler::BootstrapSampler ( const Target target,
const size_t  N 
)

Constructor.

target targets. N total number of partitions.

virtual theplu::yat::classifier::BootstrapSampler::~BootstrapSampler (  )  [virtual]

Destructor


Member Function Documentation

size_t theplu::yat::classifier::Sampler::size ( void   )  const [inherited]

Returns:
number of partitions

const Target& theplu::yat::classifier::Sampler::target ( void   )  const [inherited]

Returns:
the targets for the total set

const utility::Index& theplu::yat::classifier::Sampler::training_index ( size_t  i  )  const [inherited]

Returns:
training indices

const Target& theplu::yat::classifier::Sampler::training_target ( size_t  i  )  const [inherited]

Returns:
training target
Note:
if state is invalid the result is undefined

const utility::Index& theplu::yat::classifier::Sampler::validation_index ( size_t  i  )  const [inherited]

Returns:
validation index
Note:
if state is invalid the result is undefined

const Target& theplu::yat::classifier::Sampler::validation_target ( size_t  i  )  const [inherited]

Returns:
validation target
Note:
if state is invalid the result is undefined


The documentation for this class was generated from the following file:

Generated on Tue Jan 18 02:20:10 2011 for yat by  doxygen 1.5.5