def test_add_score_column(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
ratings_w_added_column = ri.add_score_column(
'stars', column_name='score_duplicate')
expected_columns = ['from_id', 'to_id', 'score', 'score_duplicate']
result_columns = list(ratings_w_added_column.columns)
self.assertEqual(expected_columns, result_columns)
score_column_added = list(ratings['score_duplicate'])
expected = [float(row['stars']) for row in self.raw_source_content]
self.assertEqual(expected, score_column_added)
def test_import_ratings_by_index(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
from_id_result = list(ratings['from_id'])
to_id_result = list(ratings['to_id'])
score_result = list(ratings['score'])
from_id_expected = [row['user_id'] for row in self.raw_source_content]
to_id_expected = [row['item_id'] for row in self.raw_source_content]
score_expected = [
float(row['stars']) for row in self.raw_source_content
]
self.assertTrue(
all(isinstance(from_id, str) for from_id in from_id_result))
self.assertTrue(all(isinstance(to_id, str) for to_id in to_id_result))
self.assertTrue(all(
isinstance(score, float) for score in score_result))
self.assertEqual(from_id_expected, from_id_result)
self.assertEqual(to_id_expected, to_id_result)
self.assertEqual(score_expected, score_result)
def test_add_score_column_w_score_processor(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
ratings_w_added_column = ri.add_score_column(
'review_title',
column_name='text_polarity',
score_processor=TextBlobSentimentAnalysis())
expected_columns = ['from_id', 'to_id', 'score', 'text_polarity']
result_columns = list(ratings_w_added_column.columns)
self.assertEqual(expected_columns, result_columns)
score_column_added = list(ratings['text_polarity'])
self.assertTrue(-1 <= score <= 1 for score in score_column_added)
def test_exception_add_score_column(self):
# Test exception score column can't be converted into float
ri = RatingsImporter(source=self.raw_source,
from_id_column='user_id',
to_id_column='item_id',
score_column='stars')
with self.assertRaises(ValueError):
ri.add_score_column('review_title', 'text')
def test_import_ratings_w_score_processor(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4,
score_processor=NumberNormalizer())
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
score_result = list(ratings['score'])
self.assertTrue(-1 <= score <= 1 for score in score_result)
def test_import_ratings_w_timestamp_index(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4,
timestamp_column=5)
ratings = ri.import_ratings()
expected_columns = ['from_id', 'to_id', 'score', 'timestamp']
result_columns = list(ratings.columns)
self.assertEqual(expected_columns, result_columns)
timestamp_result = list(ratings['timestamp'])
timestamp_expected = [
row['timestamp'] for row in self.raw_source_content
]
self.assertTrue(
all(isinstance(timestamp, str) for timestamp in timestamp_result))
self.assertEqual(timestamp_expected, timestamp_result)
def test_import_ratings(self):
RatingsImporter(
source=JSONFile(file_path=file_path),
output_directory="test_ratings",
rating_configs=[
RatingsFieldConfig(field_name="review_title",
processor=TextBlobSentimentAnalysis()),
RatingsFieldConfig(field_name="text",
processor=TextBlobSentimentAnalysis()),
RatingsFieldConfig(field_name="stars",
processor=NumberNormalizer(min_=0, max_=5))
],
from_field_name="user_id",
to_field_name="item_id",
timestamp_field_name="timestamp").import_ratings()
"""
def rating_config_run(config_dict: Dict):
rating_configs = []
for field in config_dict["fields"]:
class_name = field['processor'].pop('class')
class_dict = dict_detector(field["processor"])
rating_configs.append(
RatingsFieldConfig(
field_name=field["field_name"],
processor=runnable_instances[class_name](**class_dict)))
args = {}
if config_dict["source_type"] == 'sql':
pass
RatingsImporter(source=runnable_instances[config_dict["source_type"]](
file_path=config_dict["raw_source_path"], **args),
output_directory=config_dict["output_directory"],
rating_configs=rating_configs,
from_field_name=config_dict["from_field_name"],
to_field_name=config_dict["to_field_name"],
timestamp_field_name=config_dict["timestamp_field_name"]
).import_ratings()
def test_import_ratings(self):
file_path = '../../../datasets/test_import_ratings.json'
try:
with open(file_path):
pass
except FileNotFoundError:
file_path = 'datasets/test_import_ratings.json'
print(file_path)
RatingsImporter(source=JSONFile(file_path=file_path),
output_directory="test_ratings",
rating_configs=[
RatingsFieldConfig(preference_field_name="review_title",
processor=TextBlobSentimentAnalysis()),
RatingsFieldConfig(preference_field_name="text",
processor=TextBlobSentimentAnalysis()),
RatingsFieldConfig(preference_field_name="stars",
processor=NumberNormalizer(min_=0, max_=5))],
from_field_name="user_id",
to_field_name="item_id",
timestamp_field_name="timestamp").import_ratings()
# solo esempio, non presente nel dataset
"""
title_review_config = RatingsFieldConfig(
field_name='review_title',
processor=TextBlobSentimentAnalysis()
)
"""
points_review_config = RatingsFieldConfig(field_name='points',
processor=NumberNormalizer(min_=1,
max_=5))
ratings_importer = RatingsImporter(
source=CSVFile(ratings_filename), #cambia
rating_configs=[points_review_config],
from_field_name='user_id',
to_field_name='item_id',
timestamp_field_name='timestamp',
)
ratings_frame = ratings_importer.import_ratings()
print(ratings_frame)
tfidf_classifier_config = ClassifierRecommender(item_field='Plot',
field_representation='0',
classifier='random_forest')
classifier_recsys_config = RecSysConfig(
users_directory=users_ca_dir,
items_directory=items_ca_dir,
ranking_algorithm=tfidf_classifier_config,
def test_exception_import_ratings(self):
# Test exception column name not present in raw source
ri = RatingsImporter(source=self.raw_source,
from_id_column='not_existent',
to_id_column='item_id',
score_column='stars')
with self.assertRaises(KeyError):
ri.import_ratings()
# Test exception column index not present in raw source
ri = RatingsImporter(source=self.raw_source,
from_id_column=99,
to_id_column='item_id',
score_column='stars')
with self.assertRaises(IndexError):
ri.import_ratings()
# Test exception score column can't be converted into float
ri = RatingsImporter(source=self.raw_source,
from_id_column='user_id',
to_id_column='item_id',
score_column='review_title')
with self.assertRaises(ValueError):
ri.import_ratings()
def test_ratings_to_csv(self):
ri = RatingsImporter(source=self.raw_source,
from_id_column=0,
to_id_column=1,
score_column=4)
ri.import_ratings()
# Test save
ri.imported_ratings_to_csv('csv_test/')
self.assertTrue(os.path.isfile('csv_test/ratings_frame.csv'))
# Test save first duplicate
ri.imported_ratings_to_csv('csv_test/')
self.assertTrue(os.path.isfile('csv_test/ratings_frame (1).csv'))
# Test save second duplicate
ri.imported_ratings_to_csv('csv_test/')
self.assertTrue(os.path.isfile('csv_test/ratings_frame (2).csv'))
# Test save with overwrite
ri.imported_ratings_to_csv('csv_test/', overwrite=True)
self.assertTrue(os.path.isfile('csv_test/ratings_frame.csv'))
self.assertFalse(os.path.isfile('csv_test/ratings_frame (3).csv'))
# Test save with custom name
ri.imported_ratings_to_csv('csv_test/', 'ratings_custom_name')
self.assertTrue(os.path.isfile('csv_test/ratings_custom_name.csv'))
from orange_cb_recsys.evaluation import RankingAlgEvalModel, NDCG, FNMeasure, KFoldPartitioning, GiniIndex, DeltaGap, \
ReportEvalModel
from orange_cb_recsys.recsys import ClassifierRecommender, RecSysConfig, RecSys
ratings_filename = '../../../datasets/ratings_example.json'
items_ca_dir = '../../../orange_cb_recsys/movie_dir1605298315.4501655'
users_ca_dir = '../../../datasets/examples/users_dir'
stars_review_config = RatingsFieldConfig(field_name='stars',
processor=NumberNormalizer(min_=1,
max_=5))
ratings_importer = RatingsImporter(
source=JSONFile(ratings_filename),
rating_configs=[stars_review_config],
from_field_name='user_id',
to_field_name='item_id',
timestamp_field_name='timestamp',
)
ratings_frame = ratings_importer.import_ratings()
print(ratings_frame)
original_classifier_config = ClassifierRecommender(item_field='Plot',
field_representation='0',
classifier='random_forest')
classifier_recsys_config = RecSysConfig(
users_directory=users_ca_dir,
items_directory=items_ca_dir,
ranking_algorithm=original_classifier_config,
