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]))
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))
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))
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
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())
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(
Y_test[:, i], y_score[:, i])
average_precision[i] = average_precision_score(Y_test[:, i], y_score[:, i])