MyMediaLite  3.03
Public Member Functions | Properties
PosOnlyFeedbackCrossValidationSplit< T > Class Template Reference

K-fold cross-validation split for item prediction from implicit feedback. More...

Inheritance diagram for PosOnlyFeedbackCrossValidationSplit< T >:
ISplit< T >

List of all members.

Public Member Functions

 PosOnlyFeedbackCrossValidationSplit (IPosOnlyFeedback feedback, uint num_folds)
 Create a k-fold split of positive-only item prediction data.

Properties

uint NumberOfFolds [get, set]
 The number of folds in this split.
IList< IPosOnlyFeedbackTest [get, set]
 Test data for the different folds.
IList< IPosOnlyFeedbackTrain [get, set]
 Training data for the different folds.

Detailed Description

template<T>
class MyMediaLite::Data::PosOnlyFeedbackCrossValidationSplit< T >

K-fold cross-validation split for item prediction from implicit feedback.

Items with less than k events associated are ignored for testing and always assigned to the training set.

The dataset must not be modified after the split - this would lead to undefined behavior.

Type Constraints
T :IPosOnlyFeedback 
T :new() 

Member Function Documentation

PosOnlyFeedbackCrossValidationSplit ( IPosOnlyFeedback  feedback,
uint  num_folds 
) [inline]

Create a k-fold split of positive-only item prediction data.

See the class description for details.

Parameters:
feedbackthe dataset
num_foldsthe number of folds

Property Documentation

uint NumberOfFolds [get, set]

The number of folds in this split.

The number of folds in this split

Implements ISplit< T >.

IList<IPosOnlyFeedback> Test [get, set]

Test data for the different folds.

A list of T

Implements ISplit< T >.

IList<IPosOnlyFeedback> Train [get, set]

Training data for the different folds.

A list of T

Implements ISplit< T >.


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