def test_mnb_sample_weight():
clf = MultinomialNB()
clf.fit([[1, 2], [1, 2], [1, 0]], [0, 0, 1], sample_weight=[1, 1, 4])
assert_array_equal(clf.predict([[1, 0]]), [1])
positive_prior = np.exp(clf.intercept_[0])
assert_array_almost_equal([1 - positive_prior, positive_prior],
[1 / 3., 2 / 3.])
def test_check_accuracy_on_digits():
# Non regression test to make sure that any further refactoring / optim
# of the NB models do not harm the performance on a slightly non-linearly
# separable dataset
X, y = load_digits(return_X_y=True)
binary_3v8 = np.logical_or(y == 3, y == 8)
X_3v8, y_3v8 = X[binary_3v8], y[binary_3v8]
# Multinomial NB
scores = cross_val_score(MultinomialNB(alpha=10), X, y, cv=10)
assert scores.mean() > 0.86
scores = cross_val_score(MultinomialNB(alpha=10), X_3v8, y_3v8, cv=10)
assert scores.mean() > 0.94
# Bernoulli NB
scores = cross_val_score(BernoulliNB(alpha=10), X > 4, y, cv=10)
assert scores.mean() > 0.83
scores = cross_val_score(BernoulliNB(alpha=10), X_3v8 > 4, y_3v8, cv=10)
assert scores.mean() > 0.92
# Gaussian NB
scores = cross_val_score(GaussianNB(), X, y, cv=10)
assert scores.mean() > 0.77
scores = cross_val_score(GaussianNB(var_smoothing=0.1), X, y, cv=10)
assert scores.mean() > 0.89
scores = cross_val_score(GaussianNB(), X_3v8, y_3v8, cv=10)
assert scores.mean() > 0.86
def test_ovr_partial_fit():
# Test if partial_fit is working as intended
X, y = shuffle(iris.data, iris.target, random_state=0)
ovr = OneVsRestClassifier(MultinomialNB())
ovr.partial_fit(X[:100], y[:100], np.unique(y))
ovr.partial_fit(X[100:], y[100:])
pred = ovr.predict(X)
ovr2 = OneVsRestClassifier(MultinomialNB())
pred2 = ovr2.fit(X, y).predict(X)
assert_almost_equal(pred, pred2)
assert len(ovr.estimators_) == len(np.unique(y))
assert np.mean(y == pred) > 0.65
# Test when mini batches doesn't have all classes
# with SGDClassifier
X = np.abs(np.random.randn(14, 2))
y = [1, 1, 1, 1, 2, 3, 3, 0, 0, 2, 3, 1, 2, 3]
ovr = OneVsRestClassifier(
SGDClassifier(max_iter=1, tol=None, shuffle=False, random_state=0))
ovr.partial_fit(X[:7], y[:7], np.unique(y))
ovr.partial_fit(X[7:], y[7:])
pred = ovr.predict(X)
ovr1 = OneVsRestClassifier(
SGDClassifier(max_iter=1, tol=None, shuffle=False, random_state=0))
pred1 = ovr1.fit(X, y).predict(X)
assert np.mean(pred == y) == np.mean(pred1 == y)
# test partial_fit only exists if estimator has it:
ovr = OneVsRestClassifier(SVC())
assert not hasattr(ovr, "partial_fit")
def test_ovr_exceptions():
ovr = OneVsRestClassifier(LinearSVC(random_state=0))
assert_raises(ValueError, ovr.predict, [])
# Fail on multioutput data
assert_raises(ValueError,
OneVsRestClassifier(MultinomialNB()).fit,
np.array([[1, 0], [0, 1]]), np.array([[1, 2], [3, 1]]))
assert_raises(ValueError,
OneVsRestClassifier(MultinomialNB()).fit,
np.array([[1, 0], [0, 1]]), np.array([[1.5, 2.4], [3.1,
0.8]]))
def test_alpha():
# Setting alpha=0 should not output nan results when p(x_i|y_j)=0 is a case
X = np.array([[1, 0], [1, 1]])
y = np.array([0, 1])
nb = BernoulliNB(alpha=0.)
assert_warns(UserWarning, nb.partial_fit, X, y, classes=[0, 1])
assert_warns(UserWarning, nb.fit, X, y)
prob = np.array([[1, 0], [0, 1]])
assert_array_almost_equal(nb.predict_proba(X), prob)
nb = MultinomialNB(alpha=0.)
assert_warns(UserWarning, nb.partial_fit, X, y, classes=[0, 1])
assert_warns(UserWarning, nb.fit, X, y)
prob = np.array([[2. / 3, 1. / 3], [0, 1]])
assert_array_almost_equal(nb.predict_proba(X), prob)
# Test sparse X
X = scipy.sparse.csr_matrix(X)
nb = BernoulliNB(alpha=0.)
assert_warns(UserWarning, nb.fit, X, y)
prob = np.array([[1, 0], [0, 1]])
assert_array_almost_equal(nb.predict_proba(X), prob)
nb = MultinomialNB(alpha=0.)
assert_warns(UserWarning, nb.fit, X, y)
prob = np.array([[2. / 3, 1. / 3], [0, 1]])
assert_array_almost_equal(nb.predict_proba(X), prob)
# Test for alpha < 0
X = np.array([[1, 0], [1, 1]])
y = np.array([0, 1])
expected_msg = ('Smoothing parameter alpha = -1.0e-01. '
'alpha should be > 0.')
b_nb = BernoulliNB(alpha=-0.1)
m_nb = MultinomialNB(alpha=-0.1)
assert_raise_message(ValueError, expected_msg, b_nb.fit, X, y)
assert_raise_message(ValueError, expected_msg, m_nb.fit, X, y)
b_nb = BernoulliNB(alpha=-0.1)
m_nb = MultinomialNB(alpha=-0.1)
assert_raise_message(ValueError,
expected_msg,
b_nb.partial_fit,
X,
y,
classes=[0, 1])
assert_raise_message(ValueError,
expected_msg,
m_nb.partial_fit,
X,
y,
classes=[0, 1])
def test_ovo_fit_predict():
# A classifier which implements decision_function.
ovo = OneVsOneClassifier(LinearSVC(random_state=0))
ovo.fit(iris.data, iris.target).predict(iris.data)
assert len(ovo.estimators_) == n_classes * (n_classes - 1) / 2
# A classifier which implements predict_proba.
ovo = OneVsOneClassifier(MultinomialNB())
ovo.fit(iris.data, iris.target).predict(iris.data)
assert len(ovo.estimators_) == n_classes * (n_classes - 1) / 2
def test_ecoc_fit_predict():
# A classifier which implements decision_function.
ecoc = OutputCodeClassifier(LinearSVC(random_state=0),
code_size=2,
random_state=0)
ecoc.fit(iris.data, iris.target).predict(iris.data)
assert len(ecoc.estimators_) == n_classes * 2
# A classifier which implements predict_proba.
ecoc = OutputCodeClassifier(MultinomialNB(), code_size=2, random_state=0)
ecoc.fit(iris.data, iris.target).predict(iris.data)
assert len(ecoc.estimators_) == n_classes * 2
def test_ovr_multilabel():
# Toy dataset where features correspond directly to labels.
X = np.array([[0, 4, 5], [0, 5, 0], [3, 3, 3], [4, 0, 6], [6, 0, 0]])
y = np.array([[0, 1, 1], [0, 1, 0], [1, 1, 1], [1, 0, 1], [1, 0, 0]])
for base_clf in (MultinomialNB(), LinearSVC(random_state=0),
LinearRegression(), Ridge(), ElasticNet(),
Lasso(alpha=0.5)):
clf = OneVsRestClassifier(base_clf).fit(X, y)
y_pred = clf.predict([[0, 4, 4]])[0]
assert_array_equal(y_pred, [0, 1, 1])
assert clf.multilabel_
def test_ovr_partial_fit_exceptions():
ovr = OneVsRestClassifier(MultinomialNB())
X = np.abs(np.random.randn(14, 2))
y = [1, 1, 1, 1, 2, 3, 3, 0, 0, 2, 3, 1, 2, 3]
ovr.partial_fit(X[:7], y[:7], np.unique(y))
# A new class value which was not in the first call of partial_fit
# It should raise ValueError
y1 = [5] + y[7:-1]
assert_raises_regexp(ValueError, r"Mini-batch contains \[.+\] while "
r"classes must be subset of \[.+\]",
ovr.partial_fit,
X=X[7:],
y=y1)
def test_calibration_prefit():
"""Test calibration for prefitted classifiers"""
n_samples = 50
X, y = make_classification(n_samples=3 * n_samples, n_features=6,
random_state=42)
sample_weight = np.random.RandomState(seed=42).uniform(size=y.size)
X -= X.min() # MultinomialNB only allows positive X
# split train and test
X_train, y_train, sw_train = \
X[:n_samples], y[:n_samples], sample_weight[:n_samples]
X_calib, y_calib, sw_calib = \
X[n_samples:2 * n_samples], y[n_samples:2 * n_samples], \
sample_weight[n_samples:2 * n_samples]
X_test, y_test = X[2 * n_samples:], y[2 * n_samples:]
# Naive-Bayes
clf = MultinomialNB()
clf.fit(X_train, y_train, sw_train)
prob_pos_clf = clf.predict_proba(X_test)[:, 1]
# Naive Bayes with calibration
for this_X_calib, this_X_test in [(X_calib, X_test),
(sparse.csr_matrix(X_calib),
sparse.csr_matrix(X_test))]:
for method in ['isotonic', 'sigmoid']:
pc_clf = CalibratedClassifierCV(clf, method=method, cv="prefit")
for sw in [sw_calib, None]:
pc_clf.fit(this_X_calib, y_calib, sample_weight=sw)
y_prob = pc_clf.predict_proba(this_X_test)
y_pred = pc_clf.predict(this_X_test)
prob_pos_pc_clf = y_prob[:, 1]
assert_array_equal(y_pred,
np.array([0, 1])[np.argmax(y_prob, axis=1)])
assert (brier_score_loss(y_test, prob_pos_clf) >
brier_score_loss(y_test, prob_pos_pc_clf))
def test_ovr_fit_predict():
# A classifier which implements decision_function.
ovr = OneVsRestClassifier(LinearSVC(random_state=0))
pred = ovr.fit(iris.data, iris.target).predict(iris.data)
assert len(ovr.estimators_) == n_classes
clf = LinearSVC(random_state=0)
pred2 = clf.fit(iris.data, iris.target).predict(iris.data)
assert np.mean(iris.target == pred) == np.mean(iris.target == pred2)
# A classifier which implements predict_proba.
ovr = OneVsRestClassifier(MultinomialNB())
pred = ovr.fit(iris.data, iris.target).predict(iris.data)
assert np.mean(iris.target == pred) > 0.65
def test_ovo_partial_fit_predict():
temp = datasets.load_iris()
X, y = temp.data, temp.target
ovo1 = OneVsOneClassifier(MultinomialNB())
ovo1.partial_fit(X[:100], y[:100], np.unique(y))
ovo1.partial_fit(X[100:], y[100:])
pred1 = ovo1.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
ovo2.fit(X, y)
pred2 = ovo2.predict(X)
assert len(ovo1.estimators_) == n_classes * (n_classes - 1) / 2
assert np.mean(y == pred1) > 0.65
assert_almost_equal(pred1, pred2)
# Test when mini-batches have binary target classes
ovo1 = OneVsOneClassifier(MultinomialNB())
ovo1.partial_fit(X[:60], y[:60], np.unique(y))
ovo1.partial_fit(X[60:], y[60:])
pred1 = ovo1.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
pred2 = ovo2.fit(X, y).predict(X)
assert_almost_equal(pred1, pred2)
assert len(ovo1.estimators_) == len(np.unique(y))
assert np.mean(y == pred1) > 0.65
ovo = OneVsOneClassifier(MultinomialNB())
X = np.random.rand(14, 2)
y = [1, 1, 2, 3, 3, 0, 0, 4, 4, 4, 4, 4, 2, 2]
ovo.partial_fit(X[:7], y[:7], [0, 1, 2, 3, 4])
ovo.partial_fit(X[7:], y[7:])
pred = ovo.predict(X)
ovo2 = OneVsOneClassifier(MultinomialNB())
pred2 = ovo2.fit(X, y).predict(X)
assert_almost_equal(pred, pred2)
# raises error when mini-batch does not have classes from all_classes
ovo = OneVsOneClassifier(MultinomialNB())
error_y = [0, 1, 2, 3, 4, 5, 2]
message_re = escape("Mini-batch contains {0} while "
"it must be subset of {1}".format(
np.unique(error_y), np.unique(y)))
assert_raises_regexp(ValueError, message_re, ovo.partial_fit, X[:7],
error_y, np.unique(y))
# test partial_fit only exists if estimator has it:
ovr = OneVsOneClassifier(SVC())
assert not hasattr(ovr, "partial_fit")
def test_ovr_multiclass():
# Toy dataset where features correspond directly to labels.
X = np.array([[0, 0, 5], [0, 5, 0], [3, 0, 0], [0, 0, 6], [6, 0, 0]])
y = ["eggs", "spam", "ham", "eggs", "ham"]
Y = np.array([[0, 0, 1], [0, 1, 0], [1, 0, 0], [0, 0, 1], [1, 0, 0]])
classes = set("ham eggs spam".split())
for base_clf in (MultinomialNB(), LinearSVC(random_state=0),
LinearRegression(), Ridge(), ElasticNet()):
clf = OneVsRestClassifier(base_clf).fit(X, y)
assert set(clf.classes_) == classes
y_pred = clf.predict(np.array([[0, 0, 4]]))[0]
assert_array_equal(y_pred, ["eggs"])
# test input as label indicator matrix
clf = OneVsRestClassifier(base_clf).fit(X, Y)
y_pred = clf.predict([[0, 0, 4]])[0]
assert_array_equal(y_pred, [0, 0, 1])
def test_ovr_single_label_predict_proba():
base_clf = MultinomialNB(alpha=1)
X, Y = iris.data, iris.target
X_train, Y_train = X[:80], Y[:80]
X_test = X[80:]
clf = OneVsRestClassifier(base_clf).fit(X_train, Y_train)
# Decision function only estimator.
decision_only = OneVsRestClassifier(svm.SVR()).fit(X_train, Y_train)
assert not hasattr(decision_only, 'predict_proba')
Y_pred = clf.predict(X_test)
Y_proba = clf.predict_proba(X_test)
assert_almost_equal(Y_proba.sum(axis=1), 1.0)
# predict assigns a label if the probability that the
# sample has the label is greater than 0.5.
pred = np.array([l.argmax() for l in Y_proba])
assert not (pred - Y_pred).any()
def test_ovr_fit_predict_sparse():
for sparse in [
sp.csr_matrix, sp.csc_matrix, sp.coo_matrix, sp.dok_matrix,
sp.lil_matrix
]:
base_clf = MultinomialNB(alpha=1)
X, Y = datasets.make_multilabel_classification(n_samples=100,
n_features=20,
n_classes=5,
n_labels=3,
length=50,
allow_unlabeled=True,
random_state=0)
X_train, Y_train = X[:80], Y[:80]
X_test = X[80:]
clf = OneVsRestClassifier(base_clf).fit(X_train, Y_train)
Y_pred = clf.predict(X_test)
clf_sprs = OneVsRestClassifier(base_clf).fit(X_train, sparse(Y_train))
Y_pred_sprs = clf_sprs.predict(X_test)
assert clf.multilabel_
assert sp.issparse(Y_pred_sprs)
assert_array_equal(Y_pred_sprs.toarray(), Y_pred)
# Test predict_proba
Y_proba = clf_sprs.predict_proba(X_test)
# predict assigns a label if the probability that the
# sample has the label is greater than 0.5.
pred = Y_proba > .5
assert_array_equal(pred, Y_pred_sprs.toarray())
# Test decision_function
clf = svm.SVC()
clf_sprs = OneVsRestClassifier(clf).fit(X_train, sparse(Y_train))
dec_pred = (clf_sprs.decision_function(X_test) > 0).astype(int)
assert_array_equal(dec_pred, clf_sprs.predict(X_test).toarray())
def test_ovr_multilabel_predict_proba():
base_clf = MultinomialNB(alpha=1)
for au in (False, True):
X, Y = datasets.make_multilabel_classification(n_samples=100,
n_features=20,
n_classes=5,
n_labels=3,
length=50,
allow_unlabeled=au,
random_state=0)
X_train, Y_train = X[:80], Y[:80]
X_test = X[80:]
clf = OneVsRestClassifier(base_clf).fit(X_train, Y_train)
# Decision function only estimator.
decision_only = OneVsRestClassifier(svm.SVR()).fit(X_train, Y_train)
assert not hasattr(decision_only, 'predict_proba')
# Estimator with predict_proba disabled, depending on parameters.
decision_only = OneVsRestClassifier(svm.SVC(probability=False))
assert not hasattr(decision_only, 'predict_proba')
decision_only.fit(X_train, Y_train)
assert not hasattr(decision_only, 'predict_proba')
assert hasattr(decision_only, 'decision_function')
# Estimator which can get predict_proba enabled after fitting
gs = GridSearchCV(svm.SVC(probability=False),
param_grid={'probability': [True]})
proba_after_fit = OneVsRestClassifier(gs)
assert not hasattr(proba_after_fit, 'predict_proba')
proba_after_fit.fit(X_train, Y_train)
assert hasattr(proba_after_fit, 'predict_proba')
Y_pred = clf.predict(X_test)
Y_proba = clf.predict_proba(X_test)
# predict assigns a label if the probability that the
# sample has the label is greater than 0.5.
pred = Y_proba > .5
assert_array_equal(pred, Y_pred)
def test_ovr_multilabel_dataset():
base_clf = MultinomialNB(alpha=1)
for au, prec, recall in zip((True, False), (0.51, 0.66), (0.51, 0.80)):
X, Y = datasets.make_multilabel_classification(n_samples=100,
n_features=20,
n_classes=5,
n_labels=2,
length=50,
allow_unlabeled=au,
random_state=0)
X_train, Y_train = X[:80], Y[:80]
X_test, Y_test = X[80:], Y[80:]
clf = OneVsRestClassifier(base_clf).fit(X_train, Y_train)
Y_pred = clf.predict(X_test)
assert clf.multilabel_
assert_almost_equal(precision_score(Y_test, Y_pred, average="micro"),
prec,
decimal=2)
assert_almost_equal(recall_score(Y_test, Y_pred, average="micro"),
recall,
decimal=2)
def test_ovr_binary():
# Toy dataset where features correspond directly to labels.
X = np.array([[0, 0, 5], [0, 5, 0], [3, 0, 0], [0, 0, 6], [6, 0, 0]])
y = ["eggs", "spam", "spam", "eggs", "spam"]
Y = np.array([[0, 1, 1, 0, 1]]).T
classes = set("eggs spam".split())
def conduct_test(base_clf, test_predict_proba=False):
clf = OneVsRestClassifier(base_clf).fit(X, y)
assert set(clf.classes_) == classes
y_pred = clf.predict(np.array([[0, 0, 4]]))[0]
assert_array_equal(y_pred, ["eggs"])
if hasattr(base_clf, 'decision_function'):
dec = clf.decision_function(X)
assert dec.shape == (5, )
if test_predict_proba:
X_test = np.array([[0, 0, 4]])
probabilities = clf.predict_proba(X_test)
assert 2 == len(probabilities[0])
assert (clf.classes_[np.argmax(probabilities,
axis=1)] == clf.predict(X_test))
# test input as label indicator matrix
clf = OneVsRestClassifier(base_clf).fit(X, Y)
y_pred = clf.predict([[3, 0, 0]])[0]
assert y_pred == 1
for base_clf in (LinearSVC(random_state=0), LinearRegression(), Ridge(),
ElasticNet()):
conduct_test(base_clf)
for base_clf in (MultinomialNB(), SVC(probability=True),
LogisticRegression()):
conduct_test(base_clf, test_predict_proba=True)
def test_mnb_prior_unobserved_targets():
# test smoothing of prior for yet unobserved targets
# Create toy training data
X = np.array([[0, 1], [1, 0]])
y = np.array([0, 1])
clf = MultinomialNB()
assert_no_warnings(clf.partial_fit, X, y, classes=[0, 1, 2])
assert clf.predict([[0, 1]]) == 0
assert clf.predict([[1, 0]]) == 1
assert clf.predict([[1, 1]]) == 0
# add a training example with previously unobserved class
assert_no_warnings(clf.partial_fit, [[1, 1]], [2])
assert clf.predict([[0, 1]]) == 0
assert clf.predict([[1, 0]]) == 1
assert clf.predict([[1, 1]]) == 2
from mrex.ensemble import ExtraTreesClassifier
from mrex.ensemble import AdaBoostClassifier
from mrex.linear_model import LogisticRegression
from mrex.naive_bayes import MultinomialNB
ESTIMATORS = {
"dummy":
DummyClassifier(),
"random_forest":
RandomForestClassifier(max_features="sqrt", min_samples_split=10),
"extra_trees":
ExtraTreesClassifier(max_features="sqrt", min_samples_split=10),
"logistic_regression":
LogisticRegression(),
"naive_bayes":
MultinomialNB(),
"adaboost":
AdaBoostClassifier(n_estimators=10),
}
###############################################################################
# Data
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument('-e',
'--estimators',
nargs="+",
required=True,
choices=ESTIMATORS)
def test_alpha_vector():
X = np.array([[1, 0], [1, 1]])
y = np.array([0, 1])
# Setting alpha=np.array with same length
# as number of features should be fine
alpha = np.array([1, 2])
nb = MultinomialNB(alpha=alpha)
nb.partial_fit(X, y, classes=[0, 1])
# Test feature probabilities uses pseudo-counts (alpha)
feature_prob = np.array([[1 / 2, 1 / 2], [2 / 5, 3 / 5]])
assert_array_almost_equal(nb.feature_log_prob_, np.log(feature_prob))
# Test predictions
prob = np.array([[5 / 9, 4 / 9], [25 / 49, 24 / 49]])
assert_array_almost_equal(nb.predict_proba(X), prob)
# Test alpha non-negative
alpha = np.array([1., -0.1])
expected_msg = ('Smoothing parameter alpha = -1.0e-01. '
'alpha should be > 0.')
m_nb = MultinomialNB(alpha=alpha)
assert_raise_message(ValueError, expected_msg, m_nb.fit, X, y)
# Test that too small pseudo-counts are replaced
ALPHA_MIN = 1e-10
alpha = np.array([ALPHA_MIN / 2, 0.5])
m_nb = MultinomialNB(alpha=alpha)
m_nb.partial_fit(X, y, classes=[0, 1])
assert_array_almost_equal(m_nb._check_alpha(), [ALPHA_MIN, 0.5],
decimal=12)
# Test correct dimensions
alpha = np.array([1., 2., 3.])
m_nb = MultinomialNB(alpha=alpha)
expected_msg = ('alpha should be a scalar or a numpy array '
'with shape [n_features]')
assert_raise_message(ValueError, expected_msg, m_nb.fit, X, y)
def test_calibration():
"""Test calibration objects with isotonic and sigmoid"""
n_samples = 100
X, y = make_classification(n_samples=2 * n_samples, n_features=6,
random_state=42)
sample_weight = np.random.RandomState(seed=42).uniform(size=y.size)
X -= X.min() # MultinomialNB only allows positive X
# split train and test
X_train, y_train, sw_train = \
X[:n_samples], y[:n_samples], sample_weight[:n_samples]
X_test, y_test = X[n_samples:], y[n_samples:]
# Naive-Bayes
clf = MultinomialNB().fit(X_train, y_train, sample_weight=sw_train)
prob_pos_clf = clf.predict_proba(X_test)[:, 1]
pc_clf = CalibratedClassifierCV(clf, cv=y.size + 1)
assert_raises(ValueError, pc_clf.fit, X, y)
# Naive Bayes with calibration
for this_X_train, this_X_test in [(X_train, X_test),
(sparse.csr_matrix(X_train),
sparse.csr_matrix(X_test))]:
for method in ['isotonic', 'sigmoid']:
pc_clf = CalibratedClassifierCV(clf, method=method, cv=2)
# Note that this fit overwrites the fit on the entire training
# set
pc_clf.fit(this_X_train, y_train, sample_weight=sw_train)
prob_pos_pc_clf = pc_clf.predict_proba(this_X_test)[:, 1]
# Check that brier score has improved after calibration
assert (brier_score_loss(y_test, prob_pos_clf) >
brier_score_loss(y_test, prob_pos_pc_clf))
# Check invariance against relabeling [0, 1] -> [1, 2]
pc_clf.fit(this_X_train, y_train + 1, sample_weight=sw_train)
prob_pos_pc_clf_relabeled = pc_clf.predict_proba(this_X_test)[:, 1]
assert_array_almost_equal(prob_pos_pc_clf,
prob_pos_pc_clf_relabeled)
# Check invariance against relabeling [0, 1] -> [-1, 1]
pc_clf.fit(this_X_train, 2 * y_train - 1, sample_weight=sw_train)
prob_pos_pc_clf_relabeled = pc_clf.predict_proba(this_X_test)[:, 1]
assert_array_almost_equal(prob_pos_pc_clf,
prob_pos_pc_clf_relabeled)
# Check invariance against relabeling [0, 1] -> [1, 0]
pc_clf.fit(this_X_train, (y_train + 1) % 2,
sample_weight=sw_train)
prob_pos_pc_clf_relabeled = \
pc_clf.predict_proba(this_X_test)[:, 1]
if method == "sigmoid":
assert_array_almost_equal(prob_pos_pc_clf,
1 - prob_pos_pc_clf_relabeled)
else:
# Isotonic calibration is not invariant against relabeling
# but should improve in both cases
assert (brier_score_loss(y_test, prob_pos_clf) >
brier_score_loss((y_test + 1) % 2,
prob_pos_pc_clf_relabeled))
# Check failure cases:
# only "isotonic" and "sigmoid" should be accepted as methods
clf_invalid_method = CalibratedClassifierCV(clf, method="foo")
assert_raises(ValueError, clf_invalid_method.fit, X_train, y_train)
# base-estimators should provide either decision_function or
# predict_proba (most regressors, for instance, should fail)
clf_base_regressor = \
CalibratedClassifierCV(RandomForestRegressor(), method="sigmoid")
assert_raises(RuntimeError, clf_base_regressor.fit, X_train, y_train)
# Iterator over parsed Reuters SGML files.
data_stream = stream_reuters_documents()
# We learn a binary classification between the "acq" class and all the others.
# "acq" was chosen as it is more or less evenly distributed in the Reuters
# files. For other datasets, one should take care of creating a test set with
# a realistic portion of positive instances.
all_classes = np.array([0, 1])
positive_class = 'acq'
# Here are some classifiers that support the `partial_fit` method
partial_fit_classifiers = {
'SGD': SGDClassifier(max_iter=5),
'Perceptron': Perceptron(),
'NB Multinomial': MultinomialNB(alpha=0.01),
'Passive-Aggressive': PassiveAggressiveClassifier(),
}
def get_minibatch(doc_iter, size, pos_class=positive_class):
"""Extract a minibatch of examples, return a tuple X_text, y.
Note: size is before excluding invalid docs with no topics assigned.
"""
data = [('{title}\n\n{body}'.format(**doc), pos_class in doc['topics'])
for doc in itertools.islice(doc_iter, size)
if doc['topics']]
if not len(data):
return np.asarray([], dtype=int), np.asarray([], dtype=int)
def test_mnnb(kind):
# Test Multinomial Naive Bayes classification.
# This checks that MultinomialNB implements fit and predict and returns
# correct values for a simple toy dataset.
if kind == 'dense':
X = X2
elif kind == 'sparse':
X = scipy.sparse.csr_matrix(X2)
# Check the ability to predict the learning set.
clf = MultinomialNB()
assert_raises(ValueError, clf.fit, -X, y2)
y_pred = clf.fit(X, y2).predict(X)
assert_array_equal(y_pred, y2)
# Verify that np.log(clf.predict_proba(X)) gives the same results as
# clf.predict_log_proba(X)
y_pred_proba = clf.predict_proba(X)
y_pred_log_proba = clf.predict_log_proba(X)
assert_array_almost_equal(np.log(y_pred_proba), y_pred_log_proba, 8)
# Check that incremental fitting yields the same results
clf2 = MultinomialNB()
clf2.partial_fit(X[:2], y2[:2], classes=np.unique(y2))
clf2.partial_fit(X[2:5], y2[2:5])
clf2.partial_fit(X[5:], y2[5:])
y_pred2 = clf2.predict(X)
assert_array_equal(y_pred2, y2)
y_pred_proba2 = clf2.predict_proba(X)
y_pred_log_proba2 = clf2.predict_log_proba(X)
assert_array_almost_equal(np.log(y_pred_proba2), y_pred_log_proba2, 8)
assert_array_almost_equal(y_pred_proba2, y_pred_proba)
assert_array_almost_equal(y_pred_log_proba2, y_pred_log_proba)
# Partial fit on the whole data at once should be the same as fit too
clf3 = MultinomialNB()
clf3.partial_fit(X, y2, classes=np.unique(y2))
y_pred3 = clf3.predict(X)
assert_array_equal(y_pred3, y2)
y_pred_proba3 = clf3.predict_proba(X)
y_pred_log_proba3 = clf3.predict_log_proba(X)
assert_array_almost_equal(np.log(y_pred_proba3), y_pred_log_proba3, 8)
assert_array_almost_equal(y_pred_proba3, y_pred_proba)
assert_array_almost_equal(y_pred_log_proba3, y_pred_log_proba)