MyMediaLite
3.09
|
K-fold cross-validation split for item prediction from implicit feedback. More...
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< IPosOnlyFeedback > | Test [get, set] |
Test data for the different folds. | |
IList< IPosOnlyFeedback > | Train [get, set] |
Training data for the different folds. |
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.
T | : | IPosOnlyFeedback | |
T | : | new() |
PosOnlyFeedbackCrossValidationSplit | ( | IPosOnlyFeedback | feedback, |
uint | num_folds | ||
) | [inline] |
Create a k-fold split of positive-only item prediction data.
See the class description for details.
feedback | the dataset |
num_folds | the number of folds |
uint NumberOfFolds [get, set] |
IList<IPosOnlyFeedback> Test [get, set] |
IList<IPosOnlyFeedback> Train [get, set] |