def test_lipschitz_constant_loss_logreg():
rng = check_random_state(42)
X, _, w, mask = _make_data(rng=rng, masked=True)
l1_ratio = 1.
alpha = .1
grad_weight = alpha * X.shape[0] * (1. - l1_ratio)
a = _logistic_derivative_lipschitz_constant(X, mask, grad_weight)
b = _logistic_loss_lipschitz_constant(X)
assert a == b
def test_lipschitz_constant_loss_logreg():
rng = check_random_state(42)
X, _, w, mask = _make_data(rng=rng, masked=True)
l1_ratio = 1.
alpha = .1
grad_weight = alpha * X.shape[0] * (1. - l1_ratio)
a = _logistic_derivative_lipschitz_constant(X, mask, grad_weight)
b = _logistic_loss_lipschitz_constant(X)
assert_equal(a, b)
def test_logistic_lipschitz(n_samples=4, n_features=2, random_state=42):
rng = np.random.RandomState(random_state)
for scaling in np.logspace(-3, 3, num=7):
X = rng.randn(n_samples, n_features) * scaling
y = rng.randn(n_samples)
n_features = X.shape[1]
L = _logistic_loss_lipschitz_constant(X)
_check_lipschitz_continuous(lambda w: _logistic(X, y, w),
n_features + 1, L)
def test_logistic_lipschitz(n_samples=4, n_features=2, random_state=42):
rng = np.random.RandomState(random_state)
for scaling in np.logspace(-3, 3, num=7):
X = rng.randn(n_samples, n_features) * scaling
y = rng.randn(n_samples)
n_features = X.shape[1]
L = _logistic_loss_lipschitz_constant(X)
_check_lipschitz_continuous(lambda w: _logistic(
X, y, w), n_features + 1, L)