def _classify(self, sample):
"""
Predicts a sample's classification based on the training set.
Args:
sample: dict or pandas.Series
the sample or observation to be classified.
Returns:
The sample's classification.
"""
class_probabilities = self.get_classification_probabilities(sample)
return collection_utils.get_key_with_highest_value(class_probabilities)
def _classify(self, sample):
"""
Predicts a sample's classification based on the training set.
Args:
sample: dict or pandas.Series
the sample or observation to be classified.
Returns:
The sample's classification.
"""
class_probabilities = self.get_classification_probabilities(sample)
return collection_utils.get_key_with_highest_value(class_probabilities)
def choose_feature_to_split(dataset):
"""
Choose the root to be the feature which has the highest information
gain.
Args:
dataset: model.DataSet
The data set being used to build the decision tree.
Returns:
feature: string
The feature which should be the root.
"""
gains = {}
for feature in dataset.feature_list():
gains[feature] = info_gain(feature, dataset)
return get_key_with_highest_value(gains)
def choose_feature_to_split(dataset):
"""
Choose the root to be the feature which has the highest information
gain.
Args:
dataset: model.DataSet
The data set being used to build the decision tree.
Returns:
feature: string
The feature which should be the root.
"""
gains = {}
for feature in dataset.feature_list():
gains[feature] = info_gain(feature, dataset)
return get_key_with_highest_value(gains)
def _classify(self, sample):
"""
Predicts a sample's classification based on the training set.
Args:
sample:
the sample or observation to be classified.
Returns:
The sample's classification.
"""
# This function is used so that we can reduce each row with respect
# to the sample.
def calc_dist(vector):
return distance_utils.euclidean(vector, sample)
distances = self.training_set.reduce_rows(calc_dist)
votes = self._tally_votes(self.training_set.get_labels(), distances)
return collection_utils.get_key_with_highest_value(votes)
def _classify(self, sample):
"""
Predicts a sample's classification based on the training set.
Args:
sample:
the sample or observation to be classified.
Returns:
The sample's classification.
"""
# This function is used so that we can reduce each row with respect
# to the sample.
def calc_dist(vector):
return distance_utils.euclidean(vector, sample)
distances = self.training_set.reduce_rows(calc_dist)
votes = self._tally_votes(self.training_set.get_labels(), distances)
return collection_utils.get_key_with_highest_value(votes)
def test_get_key_with_highest_value_empty(self):
dictionary = {}
self.assertIsNone(
collection_utils.get_key_with_highest_value(dictionary))
def test_get_key_with_highest_value_float(self):
dictionary = {0: 0.10, 1: 0.0567, 2: 0.72}
key = collection_utils.get_key_with_highest_value(dictionary)
self.assertEqual(key, 2)
def test_get_key_with_highest_value(self):
dictionary = {"dog": 5, "cat": 10, "bird": 7}
key = collection_utils.get_key_with_highest_value(dictionary)
self.assertEqual(key, "cat")
def test_get_key_with_highest_value_empty(self):
dictionary = {}
self.assertIsNone(
collection_utils.get_key_with_highest_value(dictionary))
def test_get_key_with_highest_value_float(self):
dictionary = {0: 0.10, 1: 0.0567, 2: 0.72}
key = collection_utils.get_key_with_highest_value(dictionary)
self.assertEqual(key, 2)
def test_get_key_with_highest_value(self):
dictionary = {"dog": 5, "cat": 10, "bird": 7}
key = collection_utils.get_key_with_highest_value(dictionary)
self.assertEqual(key, "cat")
