def setUpClass(cls):
print("setUpClass")
cls.r1 = make_restaurant('A', [-10, 2], ['Fast Food', 'Thai'], 2, [
make_review('A', 4),
])
cls.r2 = make_restaurant('B', [-9, 1], ['Fast Food', 'American'], 3, [
make_review('B', 5),
make_review('B', 3.5),
])
cls.r3 = make_restaurant('C', [4, 2], [
'Fast Food',
], 1, [make_review('C', 5)])
cls.r4 = make_restaurant('D', [-2, 6], ['Sit Down', 'Thai'], 4,
[make_review('D', 2)])
cls.r5 = make_restaurant('E', [4, 2], ['Italian', 'German'], 3.5, [
make_review('E', 2.5),
make_review('E', 3),
])
cls.c1 = [0, 0]
cls.c2 = [3, 4]
cls.restaurants1 = [
make_restaurant('A', [-3, -4], [], 3, [make_review('A', 2)]),
make_restaurant('B', [1, -1], [], 1, [make_review('B', 1)]),
make_restaurant('C', [2, -4], [], 1, [make_review('C', 5)])
]
cls.restaurants2 = [
make_restaurant('D', [2, 3], [], 2, [make_review('D', 2)]),
make_restaurant('E', [0, 3], [], 3, [make_review('E', 1)])
]
def test_find_predictor(self):
user = make_user('John D.', [
make_review('A', 1),
make_review('B', 5),
make_review('C', 2),
make_review('D', 2.5),
])
restaurant = make_restaurant('New', [-10, 2], [], 2, [
make_review('New', 4),
])
cluster = [
make_restaurant('B', [4, 2], [], 1, [make_review('B', 5)]),
make_restaurant('C', [-2, 6], [], 4, [make_review('C', 2)]),
make_restaurant('D', [4, 2], [], 3.5, [
make_review('D', 2.5),
make_review('D', 3),
]),
]
pred, r_squared = find_predictor(user, cluster, restaurant_mean_rating)
val1 = round(pred(restaurant), 5)
print("val1 is: " + str(val1))
self.assertAlmostEqual(val1, 3.9359, 4)
val2 = round(r_squared, 5)
print("val2 is: " + str(val2))
self.assertAlmostEqual(val2, 0.99256, 4)
def test_find_centroid(self):
cluster1 = [
make_restaurant('A', [-3, -4], [], 3, [make_review('A', 2)]),
make_restaurant('B', [1, -1], [], 1, [make_review('B', 1)]),
make_restaurant('C', [2, -4], [], 1, [make_review('C', 5)]),
]
val = find_centroid(cluster1) # should be a pair of decimals
self.assertEqual(val, [0.0, -3.0])
def test_restaurant_ratings(self):
soda_reviews = [make_review('Soda', 4.5), make_review('Soda', 4)]
soda = make_restaurant('Soda', [127.0, 0.1],
['Restaurants', 'Breakfast & Brunch'], 1,
soda_reviews)
val = restaurant_ratings(soda)
self.assertEqual(val, [4.5, 4])
def test_restaurant_mean_rating(self):
woz_reviews = [
make_review('Wozniak Lounge', 4),
make_review('Wozniak Lounge', 3),
make_review('Wozniak Lounge', 5)
]
woz = make_restaurant('Wozniak Lounge', [127.0, 0.1],
['Restaurants', 'Pizza'], 1, woz_reviews)
val = restaurant_mean_rating(woz)
self.assertEqual(val, 4.0)
def test_best_predictor(self):
user = make_user('Cheapskate', [
make_review('A', 2),
make_review('B', 5),
make_review('C', 2),
make_review('D', 5),
])
cluster = [
make_restaurant('A', [5, 2], [], 4, [make_review('A', 5)]),
make_restaurant('B', [3, 2], [], 2, [make_review('B', 5)]),
make_restaurant('C', [-2, 6], [], 4, [make_review('C', 4)]),
]
fns = [restaurant_price, restaurant_mean_rating]
pred = best_predictor(user, cluster, fns)
print([round(pred(r), 5) for r in cluster], "SHOULD =",
[2.0, 5.0, 2.0])
self.assertEqual([round(pred(r), 5) for r in cluster], [2.0, 5.0, 2.0])
def test_rate_all(self):
user = make_user('Mr. Mean Rating Minus One', [
make_review('A', 3),
make_review('B', 4),
make_review('C', 1),
])
cluster = [
make_restaurant(
'A', [1, 2], [], 4,
[make_review('A', 4), make_review('A', 4)]),
make_restaurant('B', [4, 2], [], 3, [make_review('B', 5)]),
make_restaurant('C', [-2, 6], [], 4, [make_review('C', 2)]),
make_restaurant('D', [4, 4], [], 3.5, [
make_review('D', 2.5),
make_review('D', 3.5),
]),
]
restaurants = {restaurant_name(r): r for r in cluster}
ALL_RESTAURANTS = cluster
to_rate = cluster[2:]
fns = [restaurant_price, restaurant_mean_rating]
ratings = rate_all(user, to_rate, fns, ALL_RESTAURANTS)
print(type(ratings), "Should be ", "dict")
print(len(ratings), "Should be ", 2)
print(ratings['C'], "Should be", 1)
self.assertEqual(ratings['C'], 1)
print(round(ratings['D'], 5), "Should be ", 2.0)
self.assertEqual(ratings['D'], 2.0)