-
Notifications
You must be signed in to change notification settings - Fork 8
/
Copy pathunit_test.py
executable file
·189 lines (177 loc) · 7.89 KB
/
unit_test.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
#####################
##### UNIT TEST #####
#####################
import unittest
import numpy as np
import pandas as pd
import ranky as rk
def test_distance_matrix():
print('Distance matrix...')
m_template = pd.read_csv('data/matrix.csv')
m_template.index = m_template['index']
m_template = m_template.drop('index', axis=1).rename_axis(None, axis = 0)
print('Default')
dist_matrix = rk.distance_matrix(m_template)
print(dist_matrix)
print('Levenshtein')
dist_matrix = rk.distance_matrix(m_template, method='levenshtein')
print(dist_matrix)
def test_generator():
print('Testing generator...')
G = rk.Generator()
R = [5, 4, 3, 2, 1]
G.fit(R)
print('Judgement matrix')
m = G.sample(n=9)
print(m)
print('Score ranking')
r = rk.score(m)
print(r)
distance = rk.dist(rk.rank(R), rk.rank(r))
correlation = rk.corr(R, r)
print('hamming distance: {}'.format(distance))
print('kendall tau correlation: {}'.format(correlation))
print('Uninominal ranking')
r = rk.uninominal(m)
print(r)
distance = rk.dist(rk.rank(R), rk.rank(r))
correlation = rk.corr(R, r)
print('hamming distance: {}'.format(distance))
print('kendall tau correlation: {}'.format(correlation))
def test_metric():
""" This simply test if the metrics got computed withtout error.
It is not an unit testing.
"""
print('Testing metrics...')
y_true = pd.read_csv('data/test_metric/task.solution', sep=' ', header=None)
y_pred = pd.read_csv('data/test_metric/task.predict', sep=' ', header=None)
y_pred_proba = pd.read_csv('data/test_metric/task_proba.predict', sep=' ', header=None)
for m in ['accuracy', 'balanced_accuracy', 'precision', 'average_precision', 'f1_score', 'mxe', 'recall', 'jaccard', 'roc_auc', 'mse', 'rmse']:
try:
print('{}: {}'.format(m, rk.metric(y_true, y_pred, method=m)))
print('{}: {}'.format(m, rk.metric(y_true, y_pred_proba, method=m)))
except Exception as e:
print('Failed for {}'.format(m))
print(e)
print('Combined loss (SAR by default)')
print('{}'.format(rk.combined_metric(y_true, y_pred)))
print('{}'.format(rk.combined_metric(y_true, y_pred_proba)))
def test_utilities():
print('Testing utilities...')
leaderboard = rk.read_codalab_csv('data/chems.csv')
print(leaderboard.head())
class Test(unittest.TestCase):
M = np.array([[0.3, 0.4, 0.6], [0.8, 0.8, 0.8], [0.1, 0.5, 0.7], [0.2, 0.2, 0.2], [0, 0, 0]])
def test_rank(self):
rank_M = np.array([[2., 3., 3.], [1., 1., 1.], [4., 2., 2.], [3., 4., 4.], [5., 5., 5.]])
np.testing.assert_array_equal(rk.rank(self.__class__.M), rank_M)
def test_borda(self):
np.testing.assert_array_almost_equal(rk.borda(self.__class__.M), np.array([2.66666667, 1., 2.66666667, 3.66666667, 5.]))
m = np.array([[0.1, 0.1, 0.1], [0.3, 0.3, 0.3], [0.5, 0.5, 0.5]])
np.testing.assert_array_equal(rk.borda(m, axis=0), np.array([2, 2, 2]))
np.testing.assert_array_equal(rk.borda(m, axis=1), np.array([3, 2, 1]))
def test_majority(self):
np.testing.assert_array_equal(rk.majority(self.__class__.M), np.array([0.4, 0.8, 0.5, 0.2, 0.]))
def test_uninominal(self)
np.testing.assert_array_equal(rk.uninominal(self.__class__.M), np.array([0, 3, 0, 0, 0]))
def test_pairwise(self):
np.testing.assert_array_equal(rk.pairwise(self.__class__.M), np.array([2., 4., 3., 1., 0.]))
def test_pairwise2(self):
np.testing.assert_array_equal(rk.pairwise(self.__class__.M.T, axis=0), np.array([2., 4., 3., 1., 0.]))
def test_pairwise3(self):
np.testing.assert_array_equal(rk.pairwise(self.__class__.M, wins=rk.p_wins, pval=0.2), np.array([0., 0., 0., 0., 0.]))
def test_consensus(self):
rank_M = np.array([[1., 2., 3.], [1., 3., 2.], [1., 2., 3.]])
np.testing.assert_array_equal(rk.consensus(rank_M, axis=1), np.array([True, False, False]))
def test_winner_distance(self):
self.assertEqual(rk.dist([1, 2, 3], [2, 1, 3], method='winner'), 1)
def test_optimal_spearman_is_borda(self):
""" Check that Borda count and Spearman optimal rank aggregation returns the same output on the template matrix.
"""
m_template = pd.read_csv('data/matrix.csv')
m_template.index = m_template['index']
m_template = m_template.drop('index', axis=1).rename_axis(None, axis = 0)
borda_rank = rk.rank(rk.borda(m_template), reverse=True)
optimal_spearman_rank = rk.rank(rk.center(m_template, method='spearman'))
print(borda_rank)
print(optimal_spearman_rank)
np.testing.assert_array_equal(borda_rank, optimal_spearman_rank)
def test_kendall_w(self):
M2 = np.array([[1, 2.5, 2.5, 4], [1, 2.5, 2.5, 4], [1, 2.5, 2.5, 4]])
M3 = np.array([[2, 2, 2, 2], [2, 2, 2, 2], [2, 2, 2, 2]])
self.assertEqual(rk.kendall_w(M2), 0.9)
self.assertEqual(rk.kendall_w(M2, ties=True), 1.0)
self.assertEqual(rk.kendall_w(M3), 0.0)
def test_bayes_wins(self):
a = [0, 0, 0.2, 0, 0.3]
b = [1, 0.8, 1, 0.2, 0.2]
bw = rk.bayes_wins(a, b)
self.assertEqual(bw, False)
def test_relative_difference(self):
rd = rk.relative_difference([0, 0, 1], [0, 0, 1])
self.assertEqual(rd, 0)
rd = rk.relative_difference([0, 0], [0, 0])
self.assertEqual(rd, 0)
rd = rk.relative_difference([0.8, 0.1, 0.8], [0.2, 0.1, 0.2])
self.assertAlmostEqual(rd, 0.4)
def test_winner_distance(self):
d = rk.winner_distance([1, 0.7, 0.2, 0.1, 0.1], [0.7, 1, 0.5, 0.4, 0.1])
self.assertEqual(d, 0.25)
def test_kendall_tau_distance(self):
d = rk.kendall_tau_distance([0, 1, 2], [1, 2, 0])
self.assertEqual(d, 2)
d = rk.kendall_tau_distance([0, 1, 2], [0, 1, 2])
self.assertEqual(d, 0)
d = rk.kendall_tau_distance([0, 1, 2, 3], [0, 1, 3, 2], normalize=True)
self.assertAlmostEqual(d, 0.25)
d = rk.kendall_tau_distance([0.2, 0.2, 0.2, 0.1], [0.1, 0.2, 0.2, 0.2])
self.assertEqual(d, 4)
def test_contains_ties(self):
self.assertEqual(rk.contains_ties([0, 1, 2, 1, 2, 0]), True)
self.assertEqual(rk.contains_ties([0.4, 0.2, 0.2, 0.2]), True)
self.assertEqual(rk.contains_ties([0, 1, 2, 10, -2, 4]), False)
self.assertEqual(rk.contains_ties([0.4, 0.2, 0.1, 0.5]), False)
self.assertEqual(rk.contains_ties([]), False)
if __name__ == '__main__':
print('Compute various measures...')
m = np.array([[1, 2, 3, 4], [1, 2, 4, 3], [1, 2, 4, 3], [1, 3, 2, 4], [2, 1, 3, 4], [1, 4, 3, 2]])
print('Evolution strategy: {}'.format(rk.evolution_strategy(m, axis=1, l=5)))
print('Matrix:\n{}'.format(m))
print('Concordance: {}'.format(rk.concordance(m, axis=0)))
print('Kendall W: {}'.format(rk.kendall_w(m, axis=0)))
print('Concordance (axis=1): {}'.format(rk.concordance(m, axis=1)))
print('Kendall W (axis=1): {}'.format(rk.kendall_w(m, axis=1)))
print('Euclidean center method: {}'.format(rk.center(m, method='euclidean', axis=0)))
print('Pearson center method: {}'.format(rk.center(m, method='pearson', axis=0)))
test_generator()
test_metric()
test_utilities()
test_distance_matrix()
print('Unit testing...')
unittest.main()
'''
TODO to_binary
>>> m = np.array([[0.2, 0.3, 0.1], [0.5, 0.6, 0.7]])
>>> m
array([[0.2, 0.3, 0.1],
[0.5, 0.6, 0.7]])
>>> rk.to_binary(m)
array([[0, 0, 0],
[0, 1, 1]])
>>> rk.to_binary(m, unilabel=True)
array([[0, 1, 0],
[0, 0, 1]])
>>> rk.to_binary(m, unilabel=True, at_least_one_class=True)
array([[0, 1, 0],
[0, 0, 1]])
>>> rk.to_binary(m, unilabel=False, at_least_one_class=True)
array([[0, 1, 0],
[0, 1, 1]])
>>> m
array([[0.2, 0.3, 0.1],
[0.5, 0.6, 0.7]])
>>> rk.to_binary(m, unilabel=False, at_least_one_class=False)
array([[0, 0, 0],
[0, 1, 1]])
'''
# TODO: tests for all functions