def add_result(self, y_true, y_pred, weight=1.0):
""" Updates its statistics with the results of a prediction.
Parameters
----------
y_true: int
Index of the true label
y_pred: int
Indices of top-N predicted labels
weight: float
Sample's weight
"""
check_weights(weight)
rank = np.where(y_pred == y_true)[0]
if rank.size == 1: # Relevant item exists
self.rranks.append(1 / (rank[0] + 1)) # Works for next-item prediction
self.confusion_matrix.update(y_true, y_true, weight=weight) # True positive
else:
self.confusion_matrix.update(y_true, y_pred[0], weight=weight) # False positive
self.sample_count += 1
if self.last_true_label == y_true:
self.correct_no_change += weight
self.last_true_label = y_true
self.last_prediction = y_pred
def add_result(self, y_true, y_pred, weight=1.0):
""" Updates its statistics with the results of a prediction.
Parameters
----------
y_true: int
The true label.
y_pred: int
The classifier's prediction
weight: float
Sample's weight
"""
check_weights(weight)
true_y = self._get_target_index(y_true, True)
pred = self._get_target_index(y_pred, True)
self.confusion_matrix.update(true_y, pred)
self.sample_count += weight
if self.get_majority_class() == y_true:
self.majority_classifier = self.majority_classifier + weight
if self.last_true_label == y_true:
self.correct_no_change = self.correct_no_change + weight
self.last_true_label = y_true
self.last_prediction = y_pred
def add_result(self, y_true, y_pred, weight=1.0):
""" Updates its statistics with the results of a prediction.
If needed it will remove samples from the observation window.
Parameters
----------
y_true: int
The true label.
y_pred: int
The classifier's prediction
weight: float
Sample's weight
"""
check_weights(weight)
true_y = self._get_target_index(y_true, True)
pred = self._get_target_index(y_pred, True)
old_true = self.true_labels.add_element(np.array([y_true]))
old_predict = self.predictions.add_element(np.array([y_pred]))
# Verify if it's needed to decrease the count of any label
# pair in the confusion matrix
if (old_true is not None) and (old_predict is not None):
self.temp += 1
self.confusion_matrix.remove(
self._get_target_index(old_true[0]),
self._get_target_index(old_predict[0]))
self.correct_no_change += self.correct_no_change_correction.peek()
self.majority_classifier += self.majority_classifier_correction.peek(
)
# Verify if it's needed to decrease the majority_classifier count
if (self.get_majority_class()
== y_true) and (self.get_majority_class() is not None):
self.majority_classifier += weight
self.majority_classifier_correction.add_element([-1])
else:
self.majority_classifier_correction.add_element([0])
# Verify if it's needed to decrease the correct_no_change
if (self.last_true_label == y_true) and (self.last_true_label
is not None):
self.correct_no_change += weight
self.correct_no_change_correction.add_element([-1])
else:
self.correct_no_change_correction.add_element([0])
self.confusion_matrix.update(true_y, pred, weight=weight)
self.last_true_label = y_true
self.last_prediction = y_pred
def partial_fit(self, X, y, classes=None, sample_weight=None):
""" Partially (incrementally) fit the model.
Parameters
----------
X : numpy.ndarray of shape (n_samples, n_features)
The features to train the model.
y: numpy.ndarray of shape (n_samples)
An array-like with the class labels of all samples in X.
classes: numpy.ndarray, list, optional (default=None)
Array with all possible/known class labels. This is an optional parameter, except
for the first partial_fit call where it is compulsory.
sample_weight: numpy.ndarray of shape (n_samples), optional (default=None)
Samples weight. If not provided, uniform weights are assumed.
"""
if self.classes is None and classes is not None:
self.classes = classes
if sample_weight is None:
weight = 1.0
else:
weight = sample_weight
if y is not None:
row_cnt, _ = get_dimensions(X)
weight = check_weights(weight, expand_length=row_cnt)
for iterator in range(row_cnt):
if weight[iterator] != 0.0:
self._partial_fit(X[iterator], y[iterator], self.classes,
weight[iterator])
return self
def partial_fit(self, X, y, classes=None, weight=1.0):
if y is not None:
row_cnt, _ = get_dimensions(X)
weight = check_weights(weight, expand_length=row_cnt)
for i in range(row_cnt):
if weight[i] != 0.0:
self._train_weight_seen_by_model += weight[i]
self._partial_fit(X[i], y[i], weight[i])
def add_result(self, y_true, y_pred, weight=1.0):
""" Updates its statistics with the results of a prediction.
If needed it will remove samples from the observation window.
Parameters
----------
y_true: int
Index of the true label
y_pred: int
Indices of top-N predicted labels
weight: float
Sample's weight
"""
check_weights(weight)
old_true = self.true_labels.add_element(np.array([y_true]))
rank = np.where(y_pred == y_true)[0]
if rank.size == 1: # Relevant item exists
self.rranks.add_element(np.array([1 / (rank[0] + 1)]))
self.confusion_matrix.update(y_true, y_true, weight=weight)
old_predict = self.predictions.add_element(np.array([y_true])) # Add correct prediction
else:
self.rranks.add_element(np.array([0]))
self.confusion_matrix.update(y_true, y_pred[0], weight=weight)
old_predict = self.predictions.add_element(np.array([y_pred[0]])) # Add some incorrect prediction
# Verify if it's needed to decrease the count of any label
# pair in the confusion matrix
if (old_true is not None) and (old_predict is not None):
self.temp += 1
self.confusion_matrix.remove(old_true[0], old_predict[0])
self.correct_no_change += self.correct_no_change_correction.peek()
# Verify if it's needed to decrease the correct_no_change
if (self.last_true_label == y_true) and (self.last_true_label is not None):
self.correct_no_change += weight
self.correct_no_change_correction.add_element([-1])
else:
self.correct_no_change_correction.add_element([0])
self.last_true_label = y_true
self.last_prediction = y_pred
def test_check_weights():
test_value = 5
weights = check_weights(int(test_value))
assert weights == int(test_value)
weights = check_weights(float(test_value))
assert weights == float(test_value)
weights = check_weights(np.float(test_value))
assert weights == np.float(test_value)
weights = check_weights(np.int(test_value))
assert weights == np.int(test_value)
weights = check_weights(np.array([test_value], np.float))
assert isinstance(weights, np.ndarray)
assert isinstance(weights[0], np.float)
weights = check_weights(np.array(np.arange(test_value)))
assert isinstance(weights, np.ndarray)
for w in weights:
assert isinstance(w, np.integer)
weights = check_weights([float(x) for x in range(test_value)])
assert isinstance(weights, list)
for w in weights:
assert isinstance(w, float)
weights = check_weights(int(test_value), expand_length=10)
assert isinstance(weights, np.ndarray)
for w in weights:
assert w == int(test_value)
with pytest.raises(ValueError):
check_weights([1.0, 2.0, 3.0, 4.0, 'invalid'])
with pytest.raises(ValueError):
check_weights('invalid')
with pytest.raises(ValueError):
check_weights([float(x) for x in range(test_value)], expand_length=10)
with pytest.raises(ValueError):
check_weights(int(test_value), expand_length=-10)