def __init__(self,
train_file,
test_file,
metadata_file=None,
similarity_matrix_file=None,
prediction_file=None,
neighbors=30,
similarity_metric="correlation",
space_type='\t'):
ItemKNN.__init__(self,
train_file,
test_file,
prediction_file=prediction_file,
neighbors=neighbors,
similarity_metric=similarity_metric,
space_type=space_type)
if metadata_file is None and similarity_matrix_file is None:
print(
"This algorithm needs a similarity matrix or a metadata file!")
sys.exit(0)
if metadata_file is not None:
self.metadata = ReadFile(
metadata_file, space_type=space_type).read_metadata(self.items)
self.matrix = self.metadata['matrix'].T
self.similarity_matrix_file = similarity_matrix_file
def get_itemknn(self):
ItemKNN(train_file=self.train_file,
test_file=self.test_file,
output_file=self.output_file,
similarity_metric=self.similarity_metric,
k_neighbors=int(self.k_neighbors),
sep=self.sep,
output_sep=self.output_sep).compute(verbose_evaluation=False)
"""
# # Cross Validation
# recommender = UserKNN()
#
# CrossValidation(input_file=db, recommender=recommender, dir_folds=folds_path, header=1, k_folds=5).compute()
#
# # # Simple
# UserKNN(tr, te).compute()
# # UserAttributeKNN(tr, te, metadata_file=metadata_user).compute()
# # UserAttributeKNN(tr, te, similarity_file=sm_user).compute()
"""
ItemKNN
"""
# # Cross Validation
recommender = ItemKNN()
CrossValidation(input_file=db,
recommender=recommender,
dir_folds=folds_path,
header=1,
k_folds=5).compute()
#
# # Simple
# ItemKNN(tr, te).compute()
# ItemAttributeKNN(tr, te, metadata_file=metadata_item).compute()
# ItemAttributeKNN(tr, te, similarity_file=sm_item).compute()
def run_recommenders(self, r):
"""
1: Item KNN
2: User KNN
3: Matrix Factorization
4: SVD++
"""
flag = True
if not self.unlabeled_data[r]:
flag = False
return [], flag
else:
if r == 1:
rec = ItemKNN(self.labeled_files[r], self.unlabeled_files[r], as_similar_first=True)
rec.read_files()
rec.init_model()
rec.train_baselines()
rec.predict()
self.recommenders_predictions.setdefault(r, rec.predictions)
elif r == 2:
rec = UserKNN(self.labeled_files[r], self.unlabeled_files[r])
rec.read_files()
rec.init_model()
rec.train_baselines()
rec.predict()
self.recommenders_predictions.setdefault(r, rec.predictions)
elif r == 3:
rec = MatrixFactorization(self.labeled_files[r], self.unlabeled_files[r], random_seed=1, baseline=True)
rec.read_files()
rec.init_model()
rec.fit()
rec.predict()
self.recommenders_predictions.setdefault(r, rec.predictions)
elif r == 4:
rec = SVDPlusPlus(self.labeled_files[r], self.unlabeled_files[r], random_seed=1)
rec.read_files()
rec.fit()
rec.predict()
self.recommenders_predictions.setdefault(r, rec.predictions)
else:
raise NameError('Invalid Recommender!')
return rec.predictions, flag
CrossValidation(input_file=db,
recommender=recommender,
dir_folds=folds_path,
header=1,
k_folds=5).compute()
# # Simple
UserKNN(tr, te).compute()
# UserAttributeKNN(tr, te, metadata_file=metadata_user).compute()
# UserAttributeKNN(tr, te, similarity_file=sm_user).compute()
"""
ItemKNN
"""
# # Cross Validation
recommender = ItemKNN()
CrossValidation(input_file=db,
recommender=recommender,
dir_folds=folds_path,
header=1,
k_folds=5).compute()
#
# # Simple
ItemKNN(tr, te).compute()
# ItemAttributeKNN(tr, te, metadata_file=metadata_item).compute()
# ItemAttributeKNN(tr, te, similarity_file=sm_item).compute()
te_film,
number_sample=20,
confidence_measure='knn',
ensemble_method=True).compute() # 运行划分三个文件（每个用户预测20个对象评分）
# UserKNN(tr_film, te_film, as_similar_first=True, k_neighbors=60).compute()
# UserKNN(label_set1, te_film, as_similar_first=True, k_neighbors=60).compute()
# UserKNN(label_set2, te_film, as_similar_first=True, k_neighbors=60).compute()
# UserKNN(ensemble, te_film, as_similar_first=True, k_neighbors=60).compute()
# ItemKNN(tr_film, te_film, as_similar_first=True).compute()
ItemKNN(label_set1, te_film,
as_similar_first=True).compute() # Item_KNN以Labled_set_1为训练集的性能结果
ItemKNN(label_set2, te_film,
as_similar_first=True).compute() # Item_KNN以Labled_set_2为训练集的性能结果
ItemKNN(ensemble, te_film,
as_similar_first=True).compute() # Item_KNN以ensemble_set为训练集的性能结果
# ECoRec(tr_film, te_film, number_sample=30, confidence_measure='ec', ensemble_method=True).compute()
# ItemKNN(label_set1, te_film, as_similar_first=True).compute()
# ItemKNN(label_set2, te_film, as_similar_first=True).compute()
# ItemKNN(ensemble, te_film, as_similar_first=True).compute()