MyMediaLite item recommendation from positive-only feedback 2.02
usage: item_recommendation --training-file=FILE --recommender=METHOD [OPTIONS]
methods (plus arguments and their defaults):
- ItemAttributeSVM C=1 Gamma=0.002 (needs --item-attributes=FILE)
- SoftMarginRankingMF num_factors=10 bias_reg=0 reg_u=0.0025 reg_i=0.0025 reg_j=0.00025 num_iter=30 learn_rate=0.1 bold_driver=False fast_sampling_memory_limit=1024 init_mean=0 init_stddev=0.1
- WeightedBPRMF num_factors=10 bias_reg=0 reg_u=0.0025 reg_i=0.0025 reg_j=0.00025 num_iter=30 bold_driver=False learn_rate=0.05 init_mean=0 init_stddev=0.1
- BPRMF num_factors=10 bias_reg=0 reg_u=0.0025 reg_i=0.0025 reg_j=0.00025 num_iter=30 learn_rate=0.05 uniform_user_sampling=True with_replacement=False, bold_driver=False fast_sampling_memory_limit=1024 update_j=True init_mean=0 init_stddev=0.1
- BPR_Linear reg=0.015 num_iter=10 learn_rate=0.05 fast_sampling_memory_limit=1024 init_mean=0 init_stdev=0.1 (needs --item-attributes=FILE)
- ItemAttributeKNN k=80 (needs --item-attributes=FILE)
- ItemKNN k=80
- MostPopular
- Random
- UserAttributeKNN k=80 (needs --user-attributes=FILE)
- UserKNN k=80
- WRMF num_factors=10 regularization=0.015 c_pos=1 num_iter=15 init_mean=0 init_stdev=0.1
- WeightedItemKNN k=80
- WeightedUserKNN k=80
- Zero
method ARGUMENTS have the form name=value
general OPTIONS:
--recommender=METHOD use METHOD for recommendations (default: MostPopular)
--group-recommender=METHOD use METHOD to combine the predictions for several users
--recommender-options=OPTIONS use OPTIONS as recommender options
--help display this usage information and exit
--version display version information and exit
--random-seed=N initialize the random number generator with N
files:
--training-file=FILE read training data from FILE
--test-file=FILE read test data from FILE
--file-format=ignore_first_line|default
--data-dir=DIR load all files from DIR
--user-attributes=FILE file containing user attribute information, 1 tuple per line
--item-attributes=FILE file containing item attribute information, 1 tuple per line
--user-relations=FILE file containing user relation information, 1 tuple per line
--item-relations=FILE file containing item relation information, 1 tuple per line
--user-groups=FILE file containing group-to-user mappings, 1 tuple per line
--save-model=FILE save computed model to FILE
--load-model=FILE load model from FILE
data interpretation:
--user-prediction transpose the user-item matrix and perform user prediction instead of item prediction
--rating-threshold=NUM (for rating datasets) interpret rating >= NUM as positive feedback
choosing the items for evaluation/prediction (mutually exclusive):
--candidate-items=FILE use the items in FILE (one per line) as candidate items in the evaluation
--overlap-items use only the items that are both in the training and the test set as candidate items in the evaluation
--in-training-items use only the items in the training set as candidate items in the evaluation
--in-test-items use only the items in the test set as candidate items in the evaluation
--all-items use all known items as candidate items in the evaluation
choosing the users for evaluation/prediction
--test-users=FILE predict items for users specified in FILE (one user per line)
prediction options:
--prediction-file=FILE write ranked predictions to FILE, one user per line
--predict-items-number=N predict N items per user (needs --predict-items-file)
evaluation options:
--cross-validation=K perform k-fold cross-validation on the training data
--show-fold-results show results for individual folds in cross-validation
--test-ratio=NUM evaluate by splitting of a NUM part of the feedback
--num-test-users=N evaluate on only N randomly picked users (to save time)
--online-evaluation perform online evaluation (use every tested user-item combination for incremental training)
--filtered-evaluation perform evaluation filtered by item attribute (expects --item-attributes=FILE)
--repeat-evaluation assume that items can be accessed repeatedly - items can occur both in the training and the test data for one user
--compute-fit display fit on training data
finding the right number of iterations (iterative methods)
--find-iter=N give out statistics every N iterations
--max-iter=N perform at most N iterations
--auc-cutoff=NUM abort if AUC is below NUM
--prec5-cutoff=NUM abort if prec@5 is below NUM
[% INCLUDE footer %]