def test_gen_cities_avg(self):
# test for just one city
climate = ps5.Climate('data.csv')
test_years = pylab.array(ps5.TESTING_INTERVAL)
result = ps5.gen_cities_avg(climate, ['SEATTLE'], test_years)
correct = [11.514383561643836, 10.586849315068493, 11.28319672, 12.10643836, 12.82917808, 13.13178082]
self.assertTrue(len(correct) == len(result), "Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough, "City averages do not match expected results")
# multiple cities
result = ps5.gen_cities_avg(climate, ps5.CITIES, test_years)
correct = [16.75950424, 16.85749511, 17.56180068, 16.65717547, 16.84499022, 17.54460535]
self.assertTrue(len(correct) == len(result), "Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough, "City averages do not match expected results")
# years range
# multiple cities
result = ps5.gen_cities_avg(climate, ['TAMPA', 'DALLAS'], test_years)
correct = [20.8040411, 22.03910959, 22.27206284, 21.31136986, 20.88123288, 22.07794521]
self.assertTrue(len(correct) == len(result), "Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough, "City averages do not match expected results")
def test_gen_cities_avg(self):
# test for just one city
climate = ps5.Dataset('data.csv')
test_years = np.array(ps5.TESTING_INTERVAL)
result = ps5.gen_cities_avg(climate, ['SEATTLE'], test_years)
correct = [
10.58684932, 11.28319672, 12.10643836, 12.82917808, 13.13178082,
12.50054645
]
self.assertTrue(
len(correct) == len(result),
"Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough,
"City averages do not match expected results")
# national avg check (all cities)
result = ps5.gen_cities_avg(climate, ps5.CITIES, test_years)
correct = [
16.46957462, 17.17387834, 16.25620043, 16.47222062, 17.17817592,
17.19825999
]
self.assertTrue(
len(correct) == len(result),
"Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough,
"City averages do not match expected results")
# two-city check
result = ps5.gen_cities_avg(climate, ['TAMPA', 'DALLAS'], test_years)
correct = [
22.03910959, 22.27206284, 21.31136986, 20.88123288, 22.07794521,
22.18155738
]
self.assertTrue(
len(correct) == len(result),
"Expected length %s, was length %s" % (len(correct), len(result)))
for index in range(len(correct)):
good_enough = math.isclose(correct[index], result[index])
self.assertTrue(good_enough,
"City averages do not match expected results")
def test_find_interval(self):
# Test 1: Existing positive and negative slope intervals on city data
temp = ps5.Dataset('data.csv')
test_years = np.array(range(1961, 2016))
yearly_temps = ps5.gen_cities_avg(temp, ['PORTLAND'], test_years)
result_neg = ps5.find_interval(test_years, yearly_temps, 20, False)
correct_start = 31
correct_end = 51
self.assertIsNotNone(result_neg,
"Returned None, but valid interval exists.")
self.assertEqual(correct_start, result_neg[0], "Start year incorrect")
self.assertEqual(correct_end, result_neg[1], "End year incorrect")
result_pos = ps5.find_interval(test_years, yearly_temps, 20, True)
correct_start = 15
correct_end = 35
self.assertIsNotNone(result_pos,
"Returned None, but valid interval exists.")
self.assertEqual(correct_start, result_pos[0], "Start year incorrect")
self.assertEqual(correct_end, result_pos[1], "End year incorrect")
# Test 2: y = 2x
x = np.array(range(50))
y = np.array(range(0, 100, 2))
result_pos = ps5.find_interval(x, y, len(x) // 2, True)
result_neg = ps5.find_interval(x, y, len(x) // 2, False)
pos_correct_start = len(x) // 2
pos_correct_end = len(x)
self.assertIsNone(result_neg,
"Returned an interval, but should be None.")
self.assertIsNotNone(result_pos,
"Returned None, but valid interval exists.")
self.assertEqual(pos_correct_start, result_pos[0],
"Start year incorrect")
self.assertEqual(pos_correct_end, result_pos[1], "End year incorrect")
# Test 3: y = -2x
x = np.array(range(50))
y = np.array(range(100, 0, -2))
result_pos = ps5.find_interval(x, y, len(x) // 2, True)
result_neg = ps5.find_interval(x, y, len(x) // 2, False)
neg_correct_start = len(x) // 2
neg_correct_end = len(x)
self.assertIsNone(result_pos,
"Returned an interval, but should be None.")
self.assertIsNotNone(result_neg,
"Returned None, but valid interval exists.")
self.assertEqual(neg_correct_start, result_neg[0],
"Start year incorrect")
self.assertEqual(neg_correct_end, result_neg[1], "End year incorrect")