def test_reg_grad_data4_2(self):
y = self.data4[:, -1:]
X = self.data4[:, :-1]
m, n = X.shape
intercept = ones((m, 1), dtype=float64)
X = append(intercept, X, axis=1)
theta = -0.1 * ones((n + 1, 1), dtype=float64)
_lambda = 100
assert_allclose(array([[-0.349], [-1.710], [-49.305], [-24.727],
[-7.747], [-35.026], [-12.276], [-0.205],
[-12.948]]),
reg_grad(X, y, theta, _lambda),
rtol=0,
atol=0.001,
equal_nan=False)
def test_reg_grad_data4_3(self):
y = self.data4[:, -1:]
X = self.data4[:, :-1]
m, n = X.shape
intercept = ones((m, 1), dtype=float64)
X = append(intercept, X, axis=1)
theta = -0.1 * ones((n + 1, 1), dtype=float64)
_lambda = 1000000
assert_allclose(array([[-0.348958], [-131.906250], [-179.501291],
[-154.923177], [-137.942708], [-165.221354],
[-142.471614], [-130.400435], [-143.143226]]),
reg_grad(X, y, theta, _lambda),
rtol=0,
atol=0.001,
equal_nan=False)
def test_reg_grad_data4_1(self):
y = self.data4[:, -1:]
X = self.data4[:, :-1]
m, n = X.shape
intercept = ones((m, 1), dtype=float64)
X = append(intercept, X, axis=1)
theta = -0.1 * ones((n + 1, 1), dtype=float64)
_lambda = 10
assert_allclose(array([[-0.349], [-1.699], [-49.294], [-24.716],
[-7.736], [-35.014], [-12.265], [-0.193],
[-12.936]]),
reg_grad(X, y, theta, _lambda),
rtol=0,
atol=0.001,
equal_nan=False)
def test_reg_grad_data4_6(self):
y = self.data4[:, -1:]
X = self.data4[:, :-1]
m, n = X.shape
intercept = ones((m, 1), dtype=float64)
X = append(intercept, X, axis=1)
theta = -0.238 * ones((n + 1, 1), dtype=float64)
_lambda = 975032
def J(theta):
return reg_cost_func(X, y, theta, _lambda)
assert_allclose(reg_grad(X, y, theta, _lambda),
numerical_grad(J, theta, self.err),
rtol=0,
atol=0.001,
equal_nan=False)