def test_multivariate_normal_equation():
X = np.array([[1, 2], [2, 3], [4, 5], [6, 7], [7, 8]])
y = np.array([ 1, 2, 3, 4, 5])
X_std = (X - np.mean(X)) / X.std()
y_std = (y - np.mean(y)) / y.std()
ne_lr = LinearRegression(solver='normal_equation')
ne_lr.fit(X_std, y_std)
assert_almost_equal(ne_lr.predict(X_std), np.array(
[ -1.3054, -0.87029, 0.0000, 0.8703, 1.3054]),
decimal=4)
def test_multivariate_stochastic_gradient_descent():
X = np.array([[1, 2], [2, 3], [4, 5], [6, 7], [7, 8]])
y = np.array([ 1, 2, 3, 4, 5])
X_std = (X - np.mean(X)) / X.std()
y_std = (y - np.mean(y)) / y.std()
sgd_lr = LinearRegression(solver='sgd', eta=0.001, epochs=1000, random_seed=0)
sgd_lr.fit(X_std, y_std)
assert_almost_equal(sgd_lr.predict(X_std), np.array(
[ -1.3054, -0.87029, 0.0000, 0.8703, 1.3054]),
decimal=2)