os.environ["MKL_NUM_THREADS"] = "1"
os.environ["OPENBLAS_NUM_THREADS"] = "1"
# Data loading
data_reader = RecSys2019Reader("../data/")
data_reader = New_DataSplitter_leave_k_out(data_reader,
k_out_value=3,
use_validation_set=False,
force_new_split=True,
seed=get_split_seed())
data_reader.load_data()
URM_train, URM_test = data_reader.get_holdout_split()
# Build ICMs
ICM_numerical, _ = get_ICM_numerical(data_reader.dataReader_object)
ICM = data_reader.get_ICM_from_name("ICM_sub_class")
ICM_all, _ = merge_ICM(ICM, URM_train.transpose(), {}, {})
# Build UCMs
URM_all = data_reader.dataReader_object.get_URM_all()
UCM_age = data_reader.dataReader_object.get_UCM_from_name("UCM_age")
UCM_region = data_reader.dataReader_object.get_UCM_from_name("UCM_region")
UCM_age_region, _ = merge_UCM(UCM_age, UCM_region, {}, {})
UCM_age_region = get_warmer_UCM(UCM_age_region, URM_all, threshold_users=3)
UCM_all, _ = merge_UCM(UCM_age_region, URM_train, {}, {})
cold_items_mask = np.ediff1d(URM_train.tocsc().indptr) < 200
cold_items = np.arange(URM_train.shape[1])[cold_items_mask]
warm_users_mask = np.ediff1d(URM_train.tocsr().indptr) > 0
# Data reading
data_reader = RecSys2019Reader()
data_reader = New_DataSplitter_leave_k_out(data_reader,
k_out_value=3,
use_validation_set=False,
force_new_split=True,
seed=get_split_seed())
data_reader.load_data()
URM_train, URM_test = data_reader.get_holdout_split()
mapper = data_reader.SPLIT_GLOBAL_MAPPER_DICT['user_original_ID_to_index']
df = get_preprocessed_dataframe("../../data/", keep_warm_only=True)
# Build ICMs
ICM_numerical, _ = get_ICM_numerical(data_reader.dataReader_object)
ICM_categorical = data_reader.get_ICM_from_name("ICM_sub_class")
# Build UCMs
URM_all = data_reader.dataReader_object.get_URM_all()
UCM_age = data_reader.dataReader_object.get_UCM_from_name("UCM_age")
UCM_region = data_reader.dataReader_object.get_UCM_from_name("UCM_region")
UCM_age_region, _ = merge_UCM(UCM_age, UCM_region, {}, {})
UCM_age_region = get_warmer_UCM(UCM_age_region, URM_all, threshold_users=3)
UCM_all, _ = merge_UCM(UCM_age_region, URM_train, {}, {})
top_pop = TopPop(URM_train)
top_pop.fit()
advanced_top_pop_keywargs = {
'clustering_method': 'kmodes',