def test_multiclass_hinge_sgd():
for data in (mult_dense, mult_csr):
for fit_intercept in (True, False):
clf = SGDClassifier(loss="hinge", multiclass=True,
fit_intercept=fit_intercept, random_state=0)
clf.fit(data, mult_target)
assert_greater(clf.score(data, mult_target), 0.78)
def test_multiclass_hinge_sgd_l1l2():
for data in (mult_dense, mult_csr):
clf = SGDClassifier(loss="hinge",
penalty="l1/l2",
multiclass=True,
random_state=0)
clf.fit(data, mult_target)
assert_greater(clf.score(data, mult_target), 0.75)
def test_multiclass_hinge_sgd_l1l2(data, request):
X, y = request.getfixturevalue(data)
clf = SGDClassifier(loss="hinge",
penalty="l1/l2",
multiclass=True,
random_state=0)
clf.fit(X, y)
assert clf.score(X, y) > 0.75
def test_multiclass_hinge_sgd(data, fit_intercept, request):
X, y = request.getfixturevalue(data)
clf = SGDClassifier(loss="hinge",
multiclass=True,
fit_intercept=fit_intercept,
random_state=0)
clf.fit(X, y)
assert clf.score(X, y) > 0.78
def test_multiclass_squared_hinge_sgd():
for data in (mult_dense, mult_csr):
for fit_intercept in (True, False):
clf = SGDClassifier(loss="squared_hinge", multiclass=True,
learning_rate="constant", eta0=1e-3,
fit_intercept=fit_intercept, random_state=0)
clf.fit(data, mult_target)
assert_greater(clf.score(data, mult_target), 0.78)
def test_multiclass_squared_hinge_sgd(data, fit_intercept, request):
X, y = request.getfixturevalue(data)
clf = SGDClassifier(loss="squared_hinge",
multiclass=True,
learning_rate="constant",
eta0=1e-3,
fit_intercept=fit_intercept,
random_state=0)
clf.fit(X, y)
assert clf.score(X, y) > 0.78
def test_binary_linear_sgd():
for data in (bin_dense, bin_csr):
for clf in (
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=False,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="invscaling"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="squared_hinge",
eta0=1e-2,
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="log",
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="modified_huber",
fit_intercept=True,
learning_rate="constant"),
):
clf.fit(data, bin_target)
assert_greater(clf.score(data, bin_target), 0.934)
assert_equal(list(clf.classes_), [0, 1])
if clf.loss in ('log', 'modified_huber'):
check_predict_proba(clf, data)
else:
assert not hasattr(clf, 'predict_proba')
def test_binary_linear_sgd():
for data in (bin_dense, bin_csr):
for clf in (
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=False,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="invscaling"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="squared_hinge",
eta0=1e-2,
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="log",
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="modified_huber",
fit_intercept=True,
learning_rate="constant"),
):
clf.fit(data, bin_target)
assert_greater(clf.score(data, bin_target), 0.934)
def test_multiclass_sgd():
clf = SGDClassifier(random_state=0)
clf.fit(mult_dense, mult_target)
assert_greater(clf.score(mult_dense, mult_target), 0.80)
assert_equal(list(clf.classes_), [0, 1, 2])
def test_multiclass_hinge_sgd_l1l2():
for data in (mult_dense, mult_csr):
clf = SGDClassifier(loss="hinge", penalty="l1/l2", multiclass=True, random_state=0)
clf.fit(data, mult_target)
assert_greater(clf.score(data, mult_target), 0.75)
def test_multiclass_sgd():
clf = SGDClassifier(random_state=0)
clf.fit(mult_dense, mult_target)
assert_greater(clf.score(mult_dense, mult_target), 0.80)
def test_multiclass_sgd():
clf = SGDClassifier(random_state=0)
clf.fit(mult_dense, mult_target)
assert_greater(clf.score(mult_dense, mult_target), 0.80)
def test_multiclass_sgd():
clf = SGDClassifier(random_state=0)
clf.fit(mult_dense, mult_target)
assert clf.score(mult_dense, mult_target) > 0.80
assert list(clf.classes_) == [0, 1, 2]
def test_multiclass_sgd():
clf = SGDClassifier(random_state=0)
clf.fit(mult_dense, mult_target)
assert_greater(clf.score(mult_dense, mult_target), 0.80)
assert_equal(list(clf.classes_), [0, 1, 2])
@pytest.fixture(scope="module")
def reg_nn_train_data():
X, y, _ = make_nn_regression(n_samples=100,
n_features=10,
n_informative=8,
random_state=0)
return X, y
@pytest.mark.parametrize("data",
["bin_dense_train_data", "bin_sparse_train_data"])
@pytest.mark.parametrize("clf", [
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=False,
learning_rate="pegasos"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="invscaling"),
SGDClassifier(random_state=0,
loss="hinge",
fit_intercept=True,
learning_rate="constant"),
SGDClassifier(random_state=0,
loss="squared_hinge",
