data_splitter_holdout.hpp File Reference

Implements train/test data splitting: use initial x% of original data for training, the rest for testing. More...

#include <boost/smart_ptr.hpp>
#include <iostream>
#include <cstdlib>
#include <ctime>
#include <vector>
#include "error.hpp"
#include "global.hpp"
#include "data_splitter.hpp"

Include dependency graph for data_splitter_holdout.hpp:

Classes
class	FST::Data_Splitter_Holdout< INTERVALCONTAINER, IDXTYPE >
	Implements train/test data splitting: use initial x% of original data for training, the rest for testing, each (data)class split separately. More...

---

Detailed Description

Implements train/test data splitting: use initial x% of original data for training, the rest for testing.

====================================================================== Feature Selection Toolbox 3 source code ---------------------------------------

Author:

Petr Somol (somol@utia.cas.cz) with collaborators, see Contacts at http://fst.utia.cz

Date:

October 2010

Version:

3.0.0.beta

Note:

FST3 was developed using gcc 4.3 and requires

• Boost library (http://www.boost.org/, tested with versions 1.33.1 and 1.44),

• (optionally) LibSVM (http://www.csie.ntu.edu.tw/~cjlin/libsvm/, tested with version 3.00)

Note that LibSVM is required for SVM related tools only, as demonstrated in demo12t.cpp, demo23.cpp, demo25t.cpp, demo32t.cpp, etc.