testing.py
6c9541759d07master

testing.py
View Options

			"""
			EE-311
			======

			Lab 2: NumPy, Matplotlib and Scikit-learn (linear regression example)
			---------------------------------------------------------------------

			created by François Marelli on 03.03.2021
			"""

			import unittest
			import numpy as np
			from sklearn import datasets
			import homework

			import importlib

			importlib.reload(homework)


			class Test(unittest.TestCase):
			def test_train(self):
			expected = np.load("test_data.npy")

			wine_full, _ = datasets.load_wine(return_X_y=True)
			wine_X = wine_full[:, np.newaxis, 9]
			wine_y = wine_full[:, 0]

			data_X = wine_X[:20]
			data_y = wine_y[:20]

			predict_X = np.arange(0, 101, 10)[..., None]

			model = homework.train(data_X, data_y)
			prediction = model.predict(predict_X)

			np.testing.assert_array_almost_equal(prediction, expected)

			def test_mean_odd(self):
			array1 = np.arange(10)
			array2 = np.ones(10)
			array3 = np.arange(10) - 1
			array4 = np.ones((3, 3, 3))

			self.assertEqual(homework.mean_odd(array1), 5)
			self.assertEqual(homework.mean_odd(array2), 1)
			self.assertEqual(homework.mean_odd(array3), 3)
			self.assertEqual(homework.mean_odd(array4), 1)


			if __name__ == "__main__":
			unittest.main()