CoCalc -- public

GitHub Repository: greyhatguy007/Machine-Learning-Specialization-Coursera
Path: blob/main/C3 - Unsupervised Learning, Recommenders, Reinforcement Learning/week1/C3W1A/C3W1A2/public_tests.py
³⁵⁶⁶ views
1
import numpy as np
2
import random
3

4
def select_threshold_test(target):
5
    p_val = np.array([i / 100 for i in range(30)])
6
    y_val = np.array([1] * 5 + [0] * 25)
7
    
8
    best_epsilon, best_F1 = target(y_val, p_val)
9
    assert np.isclose(best_epsilon, 0.04, atol=0.3 / 1000), f"Wrong best_epsilon. Expected: {0.04} got: {best_epsilon}"
10
    assert best_F1 == 1, f"Wrong best_F1. Expected: 1 got: {best_F1}"
11
    
12
    y_val = np.array([1] * 5 + [0] * 25)
13
    y_val[2] = 0 # Introduce noise
14
    best_epsilon, best_F1 = target(y_val, p_val)
15
    assert np.isclose(best_epsilon, 0.04, atol=0.3 / 1000), f"Wrong best_epsilon. Expected: {0.04} got: {best_epsilon}"
16
    assert np.isclose(best_F1, 0.8888888), f"Wrong best_F1. Expected: 0.8888888 got: {best_F1}"
17
    
18
    p_val = np.array([i / 1000 for i in range(50)])
19
    y_val = np.array([1] * 8 + [0] * 42)
20
    y_val[5] = 0
21
    index = [*range(50)]
22
    random.shuffle(index)
23
    p_val = p_val[index]
24
    y_val = y_val[index]
25

26
    best_epsilon, best_F1 = target(y_val, p_val)
27
    assert np.isclose(best_epsilon, 0.007, atol=0.05 / 1000), f"Wrong best_epsilon. Expected: {0.0070070} got: {best_epsilon}"
28
    assert np.isclose(best_F1, 0.933333333), f"Wrong best_F1. Expected: 0.933333333 got: {best_F1}"
29
    print("\033[92mAll tests passed!")
30
    
31
def estimate_gaussian_test(target):
32
    np.random.seed(273)
33
    
34
    X = np.array([[1, 1, 1], 
35
                  [2, 2, 2], 
36
                  [3, 3, 3]]).T
37
    
38
    mu, var = target(X)
39
    
40
    assert type(mu) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(mu)}"
41
    assert type(var) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(var)}"
42
    
43
    assert mu.shape == (X.shape[1],), f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {mu.shape}"
44
    assert type(var) == np.ndarray, f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {var.shape}"
45
    
46
    assert np.allclose(mu, [1., 2., 3.]), f"Wrong value for mu. Expected: {[1, 2, 3]} got: {mu}"
47
    assert np.allclose(var, [0., 0., 0.]), f"Wrong value for mu. Expected: {[0, 0, 0]} got: {var}"
48
    
49
    X = np.array([[1, 2, 3], 
50
                  [2, 4, 6], 
51
                  [3, 6, 9]]).T
52
    
53
    mu, var = target(X)
54
    
55
    assert type(mu) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(mu)}"
56
    assert type(var) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(var)}"
57
    
58
    assert mu.shape == (X.shape[1],), f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {mu.shape}"
59
    assert type(var) == np.ndarray, f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {var.shape}"
60
    
61
    assert np.allclose(mu, [2., 4., 6.]), f"Wrong value for mu. Expected: {[2., 4., 6.]} got: {mu}"
62
    assert np.allclose(var, [2. / 3, 8. / 3., 18. / 3.]), f"Wrong value for mu. Expected: {[2. / 3, 8. / 3., 18. / 3.]} got: {var}"
63
    
64
    
65
    m = 500
66
    X = np.array([np.random.normal(0, 1, m), 
67
                  np.random.normal(1, 2, m), 
68
                  np.random.normal(3, 1.5, m)]).T
69
    
70
    mu, var = target(X)
71
    
72
    assert type(mu) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(mu)}"
73
    assert type(var) == np.ndarray, f"Wrong type for mu. Expected: {np.ndarray} got: {type(var)}"
74
    
75
    assert mu.shape == (X.shape[1],), f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {mu.shape}"
76
    assert type(var) == np.ndarray, f"Wrong shape for mu. Expected: {(X.shape[1],)} got: {var.shape}"
77
    
78
    assert np.allclose(mu, [0., 1., 3.], atol=0.2), f"Wrong value for mu. Expected: {[0, 1, 3]} got: {mu}"
79
    assert np.allclose(var, np.square([1., 2., 1.5]), atol=0.2), f"Wrong value for mu. Expected: {np.square([1., 2., 1.5])} got: {var}"
80
    
81
    print("\033[92mAll tests passed!")
82
Product

Resources

Company
