ItemsCrossValidation Class Reference

Cross-validation for item recommendation. More...

List of all members.

Static Public Member Functions

static Dictionary< string, double > Evaluate (ItemRecommender recommender, ISplit< IPosOnlyFeedback > split, IList< int > test_users, IList< int > candidate_items, CandidateItems candidate_item_mode=CandidateItems.OVERLAP, bool show_results=false)
 Evaluate on the folds of a dataset split.
static void EvaluateIterative (ItemRecommender recommender, ISplit< IPosOnlyFeedback > split, IList< int > test_users, IList< int > candidate_items, CandidateItems candidate_item_mode, bool repeated_events, int max_iter, int find_iter=1)
 Evaluate an iterative recommender on the folds of a dataset split, display results on STDOUT.

Detailed Description

Cross-validation for item recommendation.


Member Function Documentation

static Dictionary<string, double> Evaluate ( ItemRecommender  recommender,
ISplit< IPosOnlyFeedback split,
IList< int >  test_users,
IList< int >  candidate_items,
CandidateItems  candidate_item_mode = CandidateItems.OVERLAP,
bool  show_results = false 
) [inline, static]

Evaluate on the folds of a dataset split.

Parameters:
recommender an item recommender
split a dataset split
test_users a collection of integers with all test users
candidate_items a collection of integers with all candidate items
candidate_item_mode the mode used to determine the candidate items
show_results set to true to print results to STDERR
Returns:
a dictionary containing the average results over the different folds of the split
static void EvaluateIterative ( ItemRecommender  recommender,
ISplit< IPosOnlyFeedback split,
IList< int >  test_users,
IList< int >  candidate_items,
CandidateItems  candidate_item_mode,
bool  repeated_events,
int  max_iter,
int  find_iter = 1 
) [inline, static]

Evaluate an iterative recommender on the folds of a dataset split, display results on STDOUT.

Parameters:
recommender an item recommender
split a positive-only feedback dataset split
test_users a collection of integers with all test users
candidate_items a collection of integers with all candidate items
candidate_item_mode the mode used to determine the candidate items
repeated_events allow repeated events in the evaluation (i.e. items accessed by a user before may be in the recommended list)
max_iter the maximum number of iterations
find_iter the report interval

The documentation for this class was generated from the following file:
Generated on Sun Nov 13 20:32:48 2011 for MyMediaLite by  doxygen 1.6.3