def test_predict_raises_error_if_not_fit(self):
with self.assertRaises(PyMC3ModelsError) as no_fit_error:
test_LR = LinearRegression()
test_LR.predict(self.X_train)
expected = 'Run fit on the model before predict.'
self.assertEqual(str(no_fit_error.exception), expected)
def setUp(self):
self.num_pred = 1
self.alpha = 3
self.betas = 4
self.s = 2
# Set random seed for repeatability
np.random.seed(27)
X = np.random.randn(1000, 1)
noise = self.s * np.random.randn(1000, 1)
Y = self.betas * X + self.alpha + noise
Y = np.squeeze(Y)
self.X_train, self.X_test, self.Y_train, self.Y_test = train_test_split(
X, Y, test_size=0.4)
self.test_LR = LinearRegression()
# Fit the model with ADVI once
self.test_LR.fit(self.X_train,
self.Y_train,
num_advi_sample_draws=5000,
minibatch_size=2000)
self.nuts_LR = LinearRegression()
self.test_dir = tempfile.mkdtemp()
def test_save_and_load_work_correctly(self):
score1 = self.test_LR.score(self.X_test, self.Y_test)
self.test_LR.save(self.test_dir)
LR2 = LinearRegression()
LR2.load(self.test_dir)
self.assertEqual(self.test_LR.inference_type, LR2.inference_type)
self.assertEqual(self.test_LR.num_pred, LR2.num_pred)
self.assertEqual(self.test_LR.num_training_samples,
LR2.num_training_samples)
pd.testing.assert_frame_equal(summary(self.test_LR.trace),
summary(LR2.trace))
score2 = LR2.score(self.X_test, self.Y_test)
np.testing.assert_almost_equal(score1, score2, decimal=1)
def setUp(self):
self.num_pred = 1
self.alpha = 3
self.betas = 4
self.s = 2
X = np.random.randn(1000, 1)
noise = self.s * np.random.randn(1000, 1)
Y = self.betas * X + self.alpha + noise
Y = np.squeeze(Y)
self.X_train, self.X_test, self.Y_train, self.Y_test = train_test_split(
X, Y, test_size=0.4)
self.test_LR = LinearRegression()
self.nuts_LR = LinearRegression()
self.test_dir = tempfile.mkdtemp()