baserec.parameter_tuning package

Submodules

baserec.parameter_tuning.run_parameter_search module

@author: Maurizio Ferrari Dacrema & Ceshine Lee

baserec.parameter_tuning.run_parameter_search.run_KNNRecommender_on_similarity_type(similarity_type, parameter_searcher, parameter_search_space, recommender_input_args, n_cases, n_random_starts, resume_from_saved, save_model, evaluate_on_test, output_folder_path, output_file_name_root, metric_to_optimize, allow_weighting=False, recommender_input_args_last_test=None)

baserec.parameter_tuning.run_parameter_search.run_search_collaborative(recommender_class, URM_train, URM_train_last_test=None, n_cases=35, n_random_starts=5, resume_from_saved=False, save_model='best', evaluate_on_test='best', evaluator_validation=None, evaluator_test=None, evaluator_validation_earlystopping=None, metric_to_optimize='PRECISION', output_folder_path='result_experiments/', parallelizeKNN=True, allow_weighting=True, similarity_type_list=None)

This function performs the hyperparameter optimization for a collaborative recommender

Parameters

• recommender_class – Class of the recommender object to optimize, it must be a BaseRecommender type

• URM_train – Sparse matrix containing the URM training data

• URM_train_last_test – Sparse matrix containing the union of URM training and validation data to be used in the last evaluation

• n_cases – Number of hyperparameter sets to explore

• n_random_starts – Number of the initial random hyperparameter values to explore, usually set at 30% of n_cases

• resume_from_saved – Boolean value, if True the optimization is resumed from the saved files, if False a new one is done

• save_model – [“no”, “best”, “last”] which of the models to save, see ParameterTuning/SearchAbstractClass for details

• evaluate_on_test – [“all”, “best”, “last”, “no”] when to evaluate the model on the test data, see ParameterTuning/SearchAbstractClass for details

• evaluator_validation – Evaluator object to be used for the validation of each hyperparameter set

• evaluator_validation_earlystopping – Evaluator object to be used for the earlystopping of ML algorithms, can be the same of evaluator_validation

• evaluator_test – Evaluator object to be used for the test results, the output will only be saved but not used

• metric_to_optimize – String with the name of the metric to be optimized as contained in the output of the evaluator objects

• output_folder_path – Folder in which to save the output files

• parallelizeKNN – Boolean value, if True the various heuristics of the KNNs will be computed in parallel, if False sequentially

• allow_weighting – Boolean value, if True it enables the use of TF-IDF and BM25 to weight features, users and items in KNNs

• similarity_type_list – List of strings with the similarity heuristics to be used for the KNNs

baserec.parameter_tuning.search_abstract_class module

@author: Maurizio Ferrari Dacrema & Ceshine Lee

class baserec.parameter_tuning.search_abstract_class.SearchAbstractClass(recommender_class, evaluator_validation=None, evaluator_test=None, verbose=True)

Bases: object

ALGORITHM_NAME = 'SearchAbstractClass'

INVALID_CONFIG_VALUE = 65500.0

search(recommender_input_args, parameter_search_space, metric_to_optimize='MAP', n_cases=None, output_folder_path=None, output_file_name_root=None, parallelize=False, save_model='best', evaluate_on_test='best', save_metadata=True)

class baserec.parameter_tuning.search_abstract_class.SearchInputRecommenderArgs(CONSTRUCTOR_POSITIONAL_ARGS=None, CONSTRUCTOR_KEYWORD_ARGS=None, FIT_POSITIONAL_ARGS=None, FIT_KEYWORD_ARGS=None)

Bases: object

copy()

baserec.parameter_tuning.search_abstract_class.get_result_string_evaluate_on_validation(results_run_single_cutoff, n_decimals=7)

baserec.parameter_tuning.search_bayesian_skopt module

@author: Maurizio Ferrari Dacrema & Ceshine Lee

class baserec.parameter_tuning.search_bayesian_skopt.SearchBayesianSkopt(recommender_class, evaluator_validation=None, evaluator_test=None, verbose=True)

Bases: baserec.parameter_tuning.search_abstract_class.SearchAbstractClass

ALGORITHM_NAME = 'SearchBayesianSkopt'

search(recommender_input_args, parameter_search_space, metric_to_optimize='MAP', n_cases=20, n_random_starts=5, output_folder_path=None, output_file_name_root=None, save_model='best', save_metadata=True, resume_from_saved=False, recommender_input_args_last_test=None, evaluate_on_test='best')

Parameters

• recommender_input_args –

• parameter_search_space –

• metric_to_optimize –

• n_cases –

• n_random_starts –

• output_folder_path –

• output_file_name_root –

• save_model – “no” don’t save anything “all” save every model “best” save the best model trained on train data alone and on last, if present “last” save only last, if present

• save_metadata –

• recommender_input_args_last_test –

Returns

baserec.parameter_tuning.search_single_case module

@author: Maurizio Ferrari Dacrema & Ceshine Lee

class baserec.parameter_tuning.search_single_case.SearchSingleCase(recommender_class, evaluator_validation=None, evaluator_test=None, verbose=True)

Bases: baserec.parameter_tuning.search_abstract_class.SearchAbstractClass

ALGORITHM_NAME = 'SearchSingleCase'

search(recommender_input_args, fit_hyperparameters_values=None, metric_to_optimize='MAP', output_folder_path=None, output_file_name_root=None, save_metadata=True, recommender_input_args_last_test=None, resume_from_saved=False, save_model='best', evaluate_on_test='best')

Module contents