@@ -58,7 +58,7 @@ def test_toy_ard_object():
5858 assert_array_almost_equal (clf .predict (test ), [1 , 3 , 4 ], 2 )
5959
6060
61- def test_predict_std_bayesian ():
61+ def test_return_std_bayesian ():
6262 # generate some 1-d data with noise
6363 d = 5
6464 n_train = 50
@@ -68,6 +68,7 @@ def test_predict_std_bayesian():
6868 w = np .array ([1.0 , 0.0 , 1.0 , - 1.0 , 0.0 ])
6969 b = 1.0
7070 def f (X ): return np .dot (X , w ) + b
71+
7172 def f_noise (X ): return f (X ) + np .random .randn (X .shape [0 ])* noise_mult
7273
7374 X = np .random .random ((n_train , d ))
@@ -77,11 +78,11 @@ def f_noise(X): return f(X) + np.random.randn(X.shape[0])*noise_mult
7778 m1 = BayesianRidge ()
7879 m1 .fit (X , y )
7980 X_test = np .random .random ((n_test , d ))
80- y_mean , y_std = m1 .predict (X_test , predict_std = True )
81+ y_mean , y_std = m1 .predict (X_test , return_std = True )
8182 assert_array_almost_equal (y_std , 0.1 , decimal = 1 )
8283
8384
84- def test_predict_std_ard ():
85+ def test_return_std_ard ():
8586 # generate some 1-d data with noise
8687 d = 5
8788 n_train = 50
@@ -91,6 +92,7 @@ def test_predict_std_ard():
9192 w = np .array ([1.0 , 0.0 , 1.0 , - 1.0 , 0.0 ])
9293 b = 1.0
9394 def f (X ): return np .dot (X , w ) + b
95+
9496 def f_noise (X ): return f (X ) + np .random .randn (X .shape [0 ])* noise_mult
9597
9698 X = np .random .random ((n_train , d ))
@@ -100,5 +102,5 @@ def f_noise(X): return f(X) + np.random.randn(X.shape[0])*noise_mult
100102 m1 = ARDRegression ()
101103 m1 .fit (X , y )
102104 X_test = np .random .random ((n_test , d ))
103- y_mean , y_std = m1 .predict (X_test , predict_std = True )
105+ y_mean , y_std = m1 .predict (X_test , return_std = True )
104106 assert_array_almost_equal (y_std , 0.1 , decimal = 1 )
0 commit comments