def test_MultiTaskLassoCV():
"""Test that our MultitaskLassoCV behaves like sklearn's."""
X, y = build_dataset(n_samples=30, n_features=50, n_targets=3)
params = dict(eps=1e-2, n_alphas=10, tol=1e-10, cv=2, n_jobs=1,
fit_intercept=False, verbose=2)
clf = MultiTaskLassoCV(**params)
clf.fit(X, y)
clf2 = sklearn_MultiTaskLassoCV(**params)
clf2.max_iter = 10000 # increase max_iter bc of low tol
clf2.fit(X, y)
np.testing.assert_allclose(clf.mse_path_, clf2.mse_path_,
atol=1e-4, rtol=1e-04)
np.testing.assert_allclose(clf.alpha_, clf2.alpha_,
atol=1e-4, rtol=1e-04)
np.testing.assert_allclose(clf.coef_, clf2.coef_,
atol=1e-4, rtol=1e-04)
# check_estimator tests float32 so using tol < 1e-7 causes precision
# issues
clf.tol = 1e-5
check_estimator(clf)
def test_dropin_MultiTaskLassoCV():
"""Test that our LassoCV behaves like sklearn's LassoCV."""
X, y, _, _ = build_dataset(n_samples=30, n_features=50, n_targets=3)
params = dict(eps=1e-1,
n_alphas=100,
tol=1e-10,
cv=2,
n_jobs=2,
fit_intercept=False,
verbose=True)
clf = MultiTaskLassoCV(**params)
clf.fit(X, y)
clf2 = sklearn_MultiTaskLassoCV(**params)
clf2.fit(X, y)
np.testing.assert_allclose(clf.mse_path_, clf2.mse_path_, rtol=1e-04)
np.testing.assert_allclose(clf.alpha_, clf2.alpha_, rtol=1e-05)
np.testing.assert_allclose(clf.coef_, clf2.coef_, rtol=1e-05)
check_estimator(MultiTaskLassoCV)
n_relevant_features = 20
support = rng.choice(n_features, n_relevant_features, replace=False)
coef = np.zeros((n_tasks, n_features))
times = np.linspace(0, 2 * np.pi, n_tasks)
for k in support:
coef[:, k] = np.sin((1. + rng.randn(1)) * times + 3 * rng.randn(1))
X = rng.randn(n_samples, n_features)
Y = np.dot(X, coef.T) + rng.randn(n_samples, n_tasks)
Y /= norm(Y, ord='fro')
###############################################################################
# Fit with sklearn and celer, using the same API
params = dict(tol=1e-6, cv=4, n_jobs=-1)
t0 = time.perf_counter()
clf = MultiTaskLassoCV(**params).fit(X, Y)
t_celer = time.perf_counter() - t0
t0 = time.perf_counter()
clf_sklearn = linear_model.MultiTaskLassoCV(**params).fit(X, Y)
t_sklearn = time.perf_counter() - t0
###############################################################################
# Celer is faster
print("Time for celer : %.2f s" % t_celer)
print("Time for sklearn: %.2f s" % t_sklearn)
###############################################################################
# Both packages find the same solution
print("Celer's optimal regularizer : %s" % clf.alpha_)
print("Sklearn's optimal regularizer: %s" % clf_sklearn.alpha_)