def test_normalize(self):
data = [1.1, 2.4, 3.2, 4.6, 5.8, 6.5, 7.9, 8.1, 9.1, 10, None]
mean = 5.87
max_value = max(filter(None, data))
min_value = min(filter(None, data))
calc_result = [float(v - mean) / (max_value - min_value) for v in data if v]
calc_returned = ItemEvaluator.normalize(data)
for index, value in enumerate(calc_result):
self.assertLess(abs(calc_result[index] - calc_returned[index]), 1 / pow(10, 5))
def test_calc_more_is_better(self):
data = self.__create_test_data()
values = vector_manager.to_vector("reviews", data)
max_value = max(filter(None, values))
min_value = min(filter(None, values))
calc_result = [(v - min_value) / (max_value - min_value) for v in values if v]
calc_returned = ItemEvaluator.calc_more_is_better("reviews", data)
for index, value in enumerate(calc_result):
self.assertLess(abs(calc_result[index] - calc_returned[index]), 1 / pow(10, 5))
print(calc_result[index])
def test_calc_near_is_better(self):
data = self.__create_test_data()
values = vector_manager.to_vector("bpm", data)
max_value = max(filter(None, values))
min_value = min(filter(None, values))
selected = EvaluateItem().set_params(100, 10, datetime(2010, 4, 1, 0, 0))
calc_result = [1 - abs(v - selected.bpm) / (max_value - min_value) for v in values if v]
calc_returned = ItemEvaluator.calc_near_is_better("bpm", data, selected)
for index, value in enumerate(calc_result):
self.assertLess(abs(calc_result[index] - calc_returned[index]), 1 / pow(10, 5))
print(calc_result[index])
def test_calc_text_token_distance(self):
item_count = 5
test_items = [EvaluateItem()] * item_count
for i in range(item_count):
if i == 0:
test_items[i].tags = ["folk", "Rock", "Folk Rock", "Indie"]
elif i == 1:
test_items[i].tags = ["rock"]
elif i == 2:
test_items[i].tags = ["rock", "alternative rock"]
elif i == 3:
test_items[i].tags = ["magnet", "rock", "melodious chorus"]
elif i == 4:
test_items[i].tags = ["Rock", "Creature", "3d", "Mashup"]
distances = ItemEvaluator.calc_text_token_distance("tags", test_items, test_items[0])
print(distances)
def test_make_pattern(self):
data = self.__create_test_data()
selected = EvaluateItem().set_params(100, 10, datetime(2013, 4, 1, 0, 0))
evaluator = ItemEvaluator()
evaluator.set_rule("bpm", ItemEvaluator.calc_near_is_better)
evaluator.set_rule("reviews", ItemEvaluator.calc_more_is_better)
evaluator.set_rule("elapsed", ItemEvaluator.calc_less_is_better)
patterns = evaluator.make_pattern(data, selected)
for p in patterns:
print(p)
self.assertLess(abs(p.score - self.__calculate_pattern_percentage(p, selected, data)), 1 / pow(10, 5))
def test_get_attributes(self):
item = EvaluateItem()
attributes = ItemEvaluator.get_attributes(item)
for a in attributes:
print(a)
self.assertEquals(5, len(attributes)) # 4 is the number of attribute except private.
def make_evaluator(cls, history=None):
from .track_criticize_pattern import TrackCriticizePattern
evaluator = ItemEvaluator(TrackCriticizePattern)
evaluator.set_rule("elapsed", EvaluationType.LessIsBetter)
evaluator.set_rule("genre_score", EvaluationType.NearIsBetter)
evaluator.set_rule("bpm", EvaluationType.NearIsBetter)
evaluator.set_rule("comment_count", EvaluationType.MoreIsBetter)
evaluator.set_rule("download_count", EvaluationType.MoreIsBetter)
evaluator.set_rule("playback_count", EvaluationType.MoreIsBetter)
evaluator.set_rule("favoritings_count", EvaluationType.MoreIsBetter)
evaluator.set_rule("tag_tokens", EvaluationType.TextTokens)
# set weight by history
evaluator.set_weight("comment_count", 1.5)
evaluator.set_weight("playback_count", 2)
evaluator.set_weight("favoritings_count", 2)
return evaluator
def __get_ruled_evaluator(self):
evaluator = ItemEvaluator()
evaluator.set_rule("bpm", ItemEvaluator.calc_near_is_better)
evaluator.set_rule("reviews", ItemEvaluator.calc_more_is_better)
evaluator.set_rule("elapsed", ItemEvaluator.calc_less_is_better)
return evaluator
