def test_hat_matrix_is_symmetric(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
h_m = lr.hat_matrix
h_m_t = h_m.transpose()
assert h_m[0][0] == h_m_t[0][0]
assert h_m[0][1] == h_m_t[1][0]
def test_linear_regression(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
for i in range(1, 50):
print(lr.predict(random.uniform(1.3, 1.9)))
def test_mean_squared_error(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
mse = lr.mean_squared_error()
print("MSE : " + repr(mse))
self.assertAlmostEqual(mse, 0.01, places=2)
def test_plot(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
for i in range(1, 50):
lr.predict(random.uniform(1.3, 1.9))
lr.plot("lr.png")
def test_multiple_predictions(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
for i in range(1, 6):
lr.predict(random.uniform(1.3, 1.9))
assert len(lr.predictions) == 5
def test_predict(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
self.assertAlmostEqual(lr.predict(1.700)[1], 5.70, places=2)
def test_lr_class(self):
lr = LinearRegression(TRAINING_FILENAME, delimiter=" ")
self.assertAlmostEqual(lr.w[0], 9.49, places=2)
self.assertAlmostEqual(lr.b, -10.43, places=2)
print(lr.hat_matrix)