Example #1
0
def test_decision_function_shape_two_class():
    for n_classes in [2, 3]:
        X, y = make_blobs(centers=n_classes, random_state=0)
        for estimator in [svm.SVC, svm.NuSVC]:
            clf = OneVsRestClassifier(
                estimator(decision_function_shape="ovr")).fit(X, y)
            assert len(clf.predict(X)) == len(y)
Example #2
0
def test_ovr_single_label_decision_function():
    X, Y = datasets.make_classification(n_samples=100,
                                        n_features=20,
                                        random_state=0)
    X_train, Y_train = X[:80], Y[:80]
    X_test = X[80:]
    clf = OneVsRestClassifier(svm.SVC()).fit(X_train, Y_train)
    assert_array_equal(
        clf.decision_function(X_test).ravel() > 0, clf.predict(X_test))
Example #3
0
def test_ovr_pipeline():
    # Test with pipeline of length one
    # This test is needed because the multiclass estimators may fail to detect
    # the presence of predict_proba or decision_function.
    clf = Pipeline([("tree", DecisionTreeClassifier())])
    ovr_pipe = OneVsRestClassifier(clf)
    ovr_pipe.fit(iris.data, iris.target)
    ovr = OneVsRestClassifier(DecisionTreeClassifier())
    ovr.fit(iris.data, iris.target)
    assert_array_equal(ovr.predict(iris.data), ovr_pipe.predict(iris.data))
Example #4
0
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]]))
Example #5
0
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_
Example #6
0
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)
Example #7
0
def test_ovr_multilabel_decision_function():
    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(svm.SVC()).fit(X_train, Y_train)
    assert_array_equal((clf.decision_function(X_test) > 0).astype(int),
                       clf.predict(X_test))
Example #8
0
def plot_subfigure(X, Y, subplot, title, transform):
    if transform == "pca":
        X = PCA(n_components=2).fit_transform(X)
    elif transform == "cca":
        X = CCA(n_components=2).fit(X, Y).transform(X)
    else:
        raise ValueError

    min_x = np.min(X[:, 0])
    max_x = np.max(X[:, 0])

    min_y = np.min(X[:, 1])
    max_y = np.max(X[:, 1])

    classif = OneVsRestClassifier(SVC(kernel='linear'))
    classif.fit(X, Y)

    plt.subplot(2, 2, subplot)
    plt.title(title)

    zero_class = np.where(Y[:, 0])
    one_class = np.where(Y[:, 1])
    plt.scatter(X[:, 0], X[:, 1], s=40, c='gray', edgecolors=(0, 0, 0))
    plt.scatter(X[zero_class, 0],
                X[zero_class, 1],
                s=160,
                edgecolors='b',
                facecolors='none',
                linewidths=2,
                label='Class 1')
    plt.scatter(X[one_class, 0],
                X[one_class, 1],
                s=80,
                edgecolors='orange',
                facecolors='none',
                linewidths=2,
                label='Class 2')

    plot_hyperplane(classif.estimators_[0], min_x, max_x, 'k--',
                    'Boundary\nfor class 1')
    plot_hyperplane(classif.estimators_[1], min_x, max_x, 'k-.',
                    'Boundary\nfor class 2')
    plt.xticks(())
    plt.yticks(())

    plt.xlim(min_x - .5 * max_x, max_x + .5 * max_x)
    plt.ylim(min_y - .5 * max_y, max_y + .5 * max_y)
    if subplot == 2:
        plt.xlabel('First principal component')
        plt.ylabel('Second principal component')
        plt.legend(loc="upper left")
Example #9
0
def test_ovr_always_present():
    # Test that ovr works with classes that are always present or absent.
    # Note: tests is the case where _ConstantPredictor is utilised
    X = np.ones((10, 2))
    X[:5, :] = 0

    # Build an indicator matrix where two features are always on.
    # As list of lists, it would be: [[int(i >= 5), 2, 3] for i in range(10)]
    y = np.zeros((10, 3))
    y[5:, 0] = 1
    y[:, 1] = 1
    y[:, 2] = 1

    ovr = OneVsRestClassifier(LogisticRegression())
    assert_warns(UserWarning, ovr.fit, X, y)
    y_pred = ovr.predict(X)
    assert_array_equal(np.array(y_pred), np.array(y))
    y_pred = ovr.decision_function(X)
    assert np.unique(y_pred[:, -2:]) == 1
    y_pred = ovr.predict_proba(X)
    assert_array_equal(y_pred[:, -1], np.ones(X.shape[0]))

    # y has a constantly absent label
    y = np.zeros((10, 2))
    y[5:, 0] = 1  # variable label
    ovr = OneVsRestClassifier(LogisticRegression())
    assert_warns(UserWarning, ovr.fit, X, y)
    y_pred = ovr.predict_proba(X)
    assert_array_equal(y_pred[:, -1], np.zeros(X.shape[0]))
Example #10
0
def test_ovr_ovo_regressor():
    # test that ovr and ovo work on regressors which don't have a decision_
    # function
    ovr = OneVsRestClassifier(DecisionTreeRegressor())
    pred = ovr.fit(iris.data, iris.target).predict(iris.data)
    assert len(ovr.estimators_) == n_classes
    assert_array_equal(np.unique(pred), [0, 1, 2])
    # we are doing something sensible
    assert np.mean(pred == iris.target) > .9

    ovr = OneVsOneClassifier(DecisionTreeRegressor())
    pred = ovr.fit(iris.data, iris.target).predict(iris.data)
    assert len(ovr.estimators_) == n_classes * (n_classes - 1) / 2
    assert_array_equal(np.unique(pred), [0, 1, 2])
    # we are doing something sensible
    assert np.mean(pred == iris.target) > .9
Example #11
0
def test_ovr_gridsearch():
    ovr = OneVsRestClassifier(LinearSVC(random_state=0))
    Cs = [0.1, 0.5, 0.8]
    cv = GridSearchCV(ovr, {'estimator__C': Cs})
    cv.fit(iris.data, iris.target)
    best_C = cv.best_estimator_.estimators_[0].C
    assert best_C in Cs
Example #12
0
def test_ovr_coef_():
    for base_classifier in [
            SVC(kernel='linear', random_state=0),
            LinearSVC(random_state=0)
    ]:
        # SVC has sparse coef with sparse input data

        ovr = OneVsRestClassifier(base_classifier)
        for X in [iris.data, sp.csr_matrix(iris.data)]:
            # test with dense and sparse coef
            ovr.fit(X, iris.target)
            shape = ovr.coef_.shape
            assert shape[0] == n_classes
            assert shape[1] == iris.data.shape[1]
            # don't densify sparse coefficients
            assert (sp.issparse(ovr.estimators_[0].coef_) == sp.issparse(
                ovr.coef_))
Example #13
0
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()
Example #14
0
def test_classifier_chain_vs_independent_models():
    # Verify that an ensemble of classifier chains (each of length
    # N) can achieve a higher Jaccard similarity score than N independent
    # models
    X, Y = generate_multilabel_dataset_with_correlations()
    X_train = X[:600, :]
    X_test = X[600:, :]
    Y_train = Y[:600, :]
    Y_test = Y[600:, :]

    ovr = OneVsRestClassifier(LogisticRegression())
    ovr.fit(X_train, Y_train)
    Y_pred_ovr = ovr.predict(X_test)

    chain = ClassifierChain(LogisticRegression())
    chain.fit(X_train, Y_train)
    Y_pred_chain = chain.predict(X_test)

    assert (jaccard_score(Y_test, Y_pred_chain, average='samples') >
            jaccard_score(Y_test, Y_pred_ovr, average='samples'))
Example #15
0
def test_thresholded_scorers_multilabel_indicator_data():
    # Test that the scorer work with multilabel-indicator format
    # for multilabel and multi-output multi-class classifier
    X, y = make_multilabel_classification(allow_unlabeled=False,
                                          random_state=0)
    X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)

    # Multi-output multi-class predict_proba
    clf = DecisionTreeClassifier()
    clf.fit(X_train, y_train)
    y_proba = clf.predict_proba(X_test)
    score1 = get_scorer('roc_auc')(clf, X_test, y_test)
    score2 = roc_auc_score(y_test, np.vstack([p[:, -1] for p in y_proba]).T)
    assert_almost_equal(score1, score2)

    # Multi-output multi-class decision_function
    # TODO Is there any yet?
    clf = DecisionTreeClassifier()
    clf.fit(X_train, y_train)
    clf._predict_proba = clf.predict_proba
    clf.predict_proba = None
    clf.decision_function = lambda X: [p[:, 1] for p in clf._predict_proba(X)]

    y_proba = clf.decision_function(X_test)
    score1 = get_scorer('roc_auc')(clf, X_test, y_test)
    score2 = roc_auc_score(y_test, np.vstack([p for p in y_proba]).T)
    assert_almost_equal(score1, score2)

    # Multilabel predict_proba
    clf = OneVsRestClassifier(DecisionTreeClassifier())
    clf.fit(X_train, y_train)
    score1 = get_scorer('roc_auc')(clf, X_test, y_test)
    score2 = roc_auc_score(y_test, clf.predict_proba(X_test))
    assert_almost_equal(score1, score2)

    # Multilabel decision function
    clf = OneVsRestClassifier(LinearSVC(random_state=0))
    clf.fit(X_train, y_train)
    score1 = get_scorer('roc_auc')(clf, X_test, y_test)
    score2 = roc_auc_score(y_test, clf.decision_function(X_test))
    assert_almost_equal(score1, score2)
Example #16
0
def test_multilabel():
    """Check if error is raised for multilabel classification."""
    X, y = make_multilabel_classification(n_classes=2, n_labels=1,
                                          allow_unlabeled=False,
                                          random_state=123)
    clf = OneVsRestClassifier(SVC(kernel='linear'))

    eclf = VotingClassifier(estimators=[('ovr', clf)], voting='hard')

    try:
        eclf.fit(X, y)
    except NotImplementedError:
        return
Example #17
0
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)
Example #18
0
    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
Example #19
0
def test_ovr_coef_exceptions():
    # Not fitted exception!
    ovr = OneVsRestClassifier(LinearSVC(random_state=0))
    # lambda is needed because we don't want coef_ to be evaluated right away
    assert_raises(ValueError, lambda x: ovr.coef_, None)

    # Doesn't have coef_ exception!
    ovr = OneVsRestClassifier(DecisionTreeClassifier())
    ovr.fit(iris.data, iris.target)
    assert_raises(AttributeError, lambda x: ovr.coef_, None)
Example #20
0
def test_multiclass_multioutput_estimator():
    # test to check meta of meta estimators
    svc = LinearSVC(random_state=0)
    multi_class_svc = OneVsRestClassifier(svc)
    multi_target_svc = MultiOutputClassifier(multi_class_svc)

    multi_target_svc.fit(X, y)

    predictions = multi_target_svc.predict(X)
    assert (n_samples, n_outputs) == predictions.shape

    # train the forest with each column and assert that predictions are equal
    for i in range(3):
        multi_class_svc_ = clone(multi_class_svc)  # create a clone
        multi_class_svc_.fit(X, y[:, i])
        assert (list(multi_class_svc_.predict(X)) == list(predictions[:, i]))
Example #21
0
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
Example #22
0
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])
Example #23
0
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")
Example #24
0
y = label_binarize(y, classes=[0, 1, 2])
n_classes = y.shape[1]

# Add noisy features to make the problem harder
random_state = np.random.RandomState(0)
n_samples, n_features = X.shape
X = np.c_[X, random_state.randn(n_samples, 200 * n_features)]

# shuffle and split training and test sets
X_train, X_test, y_train, y_test = train_test_split(X,
                                                    y,
                                                    test_size=.5,
                                                    random_state=0)

# Learn to predict each class against the other
classifier = OneVsRestClassifier(
    svm.SVC(kernel='linear', probability=True, random_state=random_state))
y_score = classifier.fit(X_train, y_train).decision_function(X_test)

# Compute ROC curve and ROC area for each class
fpr = dict()
tpr = dict()
roc_auc = dict()
for i in range(n_classes):
    fpr[i], tpr[i], _ = roc_curve(y_test[:, i], y_score[:, i])
    roc_auc[i] = auc(fpr[i], tpr[i])

# Compute micro-average ROC curve and ROC area
fpr["micro"], tpr["micro"], _ = roc_curve(y_test.ravel(), y_score.ravel())
roc_auc["micro"] = auc(fpr["micro"], tpr["micro"])

##############################################################################
Example #25
0
from mrex.linear_model import LogisticRegression

print(__doc__)

# Load a multi-label dataset from https://www.openml.org/d/40597
X, Y = fetch_openml('yeast', version=4, return_X_y=True)
Y = Y == 'TRUE'
X_train, X_test, Y_train, Y_test = train_test_split(X,
                                                    Y,
                                                    test_size=.2,
                                                    random_state=0)

# Fit an independent logistic regression model for each class using the
# OneVsRestClassifier wrapper.
base_lr = LogisticRegression()
ovr = OneVsRestClassifier(base_lr)
ovr.fit(X_train, Y_train)
Y_pred_ovr = ovr.predict(X_test)
ovr_jaccard_score = jaccard_score(Y_test, Y_pred_ovr, average='samples')

# Fit an ensemble of logistic regression classifier chains and take the
# take the average prediction of all the chains.
chains = [
    ClassifierChain(base_lr, order='random', random_state=i) for i in range(10)
]
for chain in chains:
    chain.fit(X_train, Y_train)

Y_pred_chains = np.array([chain.predict(X_test) for chain in chains])
chain_jaccard_scores = [
    jaccard_score(Y_test, Y_pred_chain >= .5, average='samples')
Example #26
0
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())
Example #27
0
# Use label_binarize to be multi-label like settings
Y = label_binarize(y, classes=[0, 1, 2])
n_classes = Y.shape[1]

# Split into training and test
X_train, X_test, Y_train, Y_test = train_test_split(X,
                                                    Y,
                                                    test_size=.5,
                                                    random_state=random_state)

# We use OneVsRestClassifier for multi-label prediction
from mrex.multiclass import OneVsRestClassifier

# Run classifier
classifier = OneVsRestClassifier(svm.LinearSVC(random_state=random_state))
classifier.fit(X_train, Y_train)
y_score = classifier.decision_function(X_test)

###############################################################################
# The average precision score in multi-label settings
# ....................................................
from mrex.metrics import precision_recall_curve
from mrex.metrics import average_precision_score

# For each class
precision = dict()
recall = dict()
average_precision = dict()
for i in range(n_classes):
    precision[i], recall[i], _ = precision_recall_curve(
Example #28
0
def test_ovr_fit_predict_svc():
    ovr = OneVsRestClassifier(svm.SVC())
    ovr.fit(iris.data, iris.target)
    assert len(ovr.estimators_) == 3
    assert ovr.score(iris.data, iris.target) > .9
Example #29
0
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)