File size: 11,501 Bytes
fc0f7bd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
# Copyright (c) Microsoft Corporation. All rights reserved.
# Licensed under the MIT License.

import numpy as np
import pytest

import fairlearn.metrics as metrics
from test.unit.input_convertors import conversions_for_1d


# ===========================================================


def mock_func(y_true, y_pred):
    return np.sum(y_true)


def mock_func_weight(y_true, y_pred, sample_weight):
    return np.sum(np.multiply(y_true, sample_weight))


def mock_func_matrix_return(y_true, y_pred):
    return np.ones([len(y_true), sum(y_pred)])


class TestMetricByGroup:
    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_smoke(self, transform_y_a, transform_y_p, transform_gid):
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0, 1])
        gid = transform_gid([0, 0, 0, 0, 1, 1, 1, 1])

        result = metrics.metric_by_group(mock_func, y_a, y_p, gid)

        assert result.overall == 5
        assert len(result.by_group) == 2
        assert result.by_group[0] == 2
        assert result.by_group[1] == 3
        assert result.minimum == 2
        assert result.argmin_set == {0}
        assert result.maximum == 3
        assert result.argmax_set == {1}
        assert result.range == 1
        assert result.range_ratio == pytest.approx(0.6666666667)

    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_string_groups(self, transform_y_a, transform_y_p, transform_gid):
        a = "ABC"
        b = "DEF"
        c = "GHI"
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0, 1])
        gid = transform_gid([a, a, a, b, b, c, c, c])

        result = metrics.metric_by_group(mock_func, y_a, y_p, gid)

        assert result.overall == 5
        assert len(result.by_group) == 3
        assert result.by_group[a] == 1
        assert result.by_group[b] == 1
        assert result.by_group[c] == 3
        assert result.minimum == 1
        assert result.argmin_set == {a, b}
        assert result.maximum == 3
        assert result.argmax_set == {c}
        assert result.range == 2
        assert result.range_ratio == pytest.approx(0.33333333333333)

    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_matrix_metric(self, transform_y_a, transform_y_p, transform_gid):
        a = "ABC"
        b = "DEF"
        c = "GHI"
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0, 1])
        gid = transform_gid([a, a, a, b, b, c, c, c])

        result = metrics.metric_by_group(mock_func_matrix_return, y_a, y_p, gid)

        assert np.array_equal(result.overall, np.ones([8, 5]))
        assert np.array_equal(result.by_group[a], np.ones([3, 2]))
        assert np.array_equal(result.by_group[b], np.ones([2, 2]))
        assert np.array_equal(result.by_group[c], np.ones([3, 1]))

    def test_matrix_metric_other_properties(self):
        a = "ABC"
        b = "DEF"
        c = "GHI"
        y_a = [0, 0, 1, 1, 0, 1, 1, 1]
        y_p = [0, 1, 1, 1, 1, 0, 0, 1]
        gid = [a, a, a, b, b, c, c, c]

        result = metrics.metric_by_group(mock_func_matrix_return, y_a, y_p, gid)

        # Other fields should fail
        with pytest.raises(ValueError):
            _ = result.minimum
        with pytest.raises(ValueError):
            _ = result.maximum
        with pytest.raises(ValueError):
            _ = result.argmin_set
        with pytest.raises(ValueError):
            _ = result.argmax_set
        with pytest.raises(ValueError):
            _ = result.range
        with pytest.raises(ValueError):
            _ = result.range_ratio

    @pytest.mark.parametrize("transform_s_w", conversions_for_1d)
    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_with_weights(self, transform_y_a, transform_y_p, transform_gid, transform_s_w):
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0, 1])
        gid = transform_gid([0, 0, 0, 0, 1, 1, 2, 2])
        s_w = transform_s_w([1, 1, 1, 1, 2, 2, 3, 3])

        result = metrics.metric_by_group(mock_func_weight, y_a, y_p, gid, sample_weight=s_w)

        assert result.overall == 10
        assert len(result.by_group) == 3
        assert result.by_group[0] == 2
        assert result.by_group[1] == 2
        assert result.by_group[2] == 6
        assert result.minimum == 2
        assert result.argmin_set == {0, 1}
        assert result.maximum == 6
        assert result.argmax_set == {2}
        assert result.range == 4
        assert result.range_ratio == pytest.approx(0.33333333333333)

    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_true_predict_length_mismatch(self, transform_y_a, transform_y_p):
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0])
        gid = [0, 0, 0, 0, 1, 1, 2, 2]
        s_w = [1, 1, 1, 1, 2, 2, 3, 3]

        with pytest.raises(ValueError) as exception_context:
            _ = metrics.metric_by_group(mock_func_weight, y_a, y_p, gid, s_w)

        expected = "Array y_pred is not the same size as y_true"
        assert exception_context.value.args[0] == expected

    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_true_group_length_mismatch(self, transform_y_a, transform_gid):
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = [0, 1, 1, 1, 1, 0, 0, 0]
        gid = transform_gid([0, 0, 0, 0, 1, 1, 2])
        s_w = [1, 1, 1, 1, 2, 2, 3, 3]

        with pytest.raises(ValueError) as exception_context:
            _ = metrics.metric_by_group(mock_func_weight, y_a, y_p, gid, s_w)

        expected = "Array group_membership is not the same size as y_true"
        assert exception_context.value.args[0] == expected

    @pytest.mark.parametrize("transform_s_w", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_true_weight_length_mismatch(self, transform_y_a, transform_s_w):
        y_a = transform_y_a([0, 0, 1, 1, 0, 1, 1, 1])
        y_p = [0, 1, 1, 1, 1, 0, 0, 0]
        gid = [0, 0, 0, 0, 1, 1, 2, 3]
        s_w = transform_s_w([1, 1, 1, 1, 2, 2, 3])

        with pytest.raises(ValueError) as exception_context:
            _ = metrics.metric_by_group(mock_func_weight, y_a, y_p, gid, s_w)

        expected = "Array sample_weight is not the same size as y_true"
        assert exception_context.value.args[0] == expected

    def test_negative_results(self):
        y_a = [0, 0, 1, 1, 0, 1, 1, 1]
        y_p = [0, 1, 1, 1, 1, 0, 0, 1]
        gid = [0, 0, 0, 0, 0, 1, 1, 1]

        def negative_results(y_true, y_pred):
            return -(len(y_true) + len(y_pred))

        result = metrics.metric_by_group(negative_results, y_a, y_p, gid)

        assert result.overall == -16
        assert result.by_group[0] == -10
        assert result.by_group[1] == -6
        assert result.minimum == -10
        assert result.maximum == -6
        assert result.range == 4
        assert np.isnan(result.range_ratio)

    def test_metric_results_zero(self):
        y_a = [0, 0, 1, 1, 0, 1, 1, 1]
        y_p = [0, 1, 1, 1, 1, 0, 0, 1]
        gid = [0, 0, 0, 0, 0, 1, 1, 1]

        def zero_results(y_true, y_pred):
            # Arrays will always be same length
            return len(y_true)-len(y_pred)

        result = metrics.metric_by_group(zero_results, y_a, y_p, gid)

        assert result.overall == 0
        assert result.by_group[0] == 0
        assert result.by_group[1] == 0
        assert result.minimum == 0
        assert result.maximum == 0
        assert result.range == 0
        # Following is special case
        assert result.range_ratio == 1

    def test_single_element_input(self):
        y_t = [0]
        y_p = [0]
        gid = [0]
        s_w = [0]

        def sum_lengths(y_true, y_pred, sample_weight):
            return len(y_true) + len(y_pred) + len(sample_weight)

        result = metrics.metric_by_group(sum_lengths, y_t, y_p, gid, sample_weight=s_w)
        assert result.overall == 3
        assert result.by_group[0] == 3
        assert result.minimum == 3
        assert result.maximum == 3
        assert result.range == 0
        assert result.range_ratio == 1

    def test_groups_only_one_element(self):
        y_t = [1, 2]
        y_p = [1, 2]
        gid = [0, 1]

        def sum_lengths(y_true, y_pred):
            return len(y_true) + len(y_pred)

        result = metrics.metric_by_group(sum_lengths, y_t, y_p, gid)
        assert result.overall == 4
        assert result.by_group[0] == 2
        assert result.by_group[1] == 2
        assert result.minimum == 2
        assert result.maximum == 2
        assert result.range == 0
        assert result.range_ratio == 1


class TestMakeGroupMetric:
    def test_smoke(self):
        y_a = [0, 0, 1, 1, 0, 1, 1, 1]
        y_p = [0, 1, 1, 1, 1, 0, 0, 1]
        gid = [0, 0, 0, 0, 1, 1, 1, 1]

        grouped_metric_func = metrics.make_group_metric(mock_func)
        result = grouped_metric_func(y_a, y_p, gid)
        assert result.overall == 5
        assert len(result.by_group) == 2
        assert result.by_group[0] == 2
        assert result.by_group[1] == 3
        assert result.minimum == 2
        assert result.maximum == 3
        assert result.argmin_set == {0}
        assert result.argmax_set == {1}
        assert result.range == 1
        assert result.range_ratio == pytest.approx(0.66666666667)

    @pytest.mark.parametrize("transform_s_w", conversions_for_1d)
    @pytest.mark.parametrize("transform_gid", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_p", conversions_for_1d)
    @pytest.mark.parametrize("transform_y_a", conversions_for_1d)
    def test_keys_and_weights(self, transform_y_a, transform_y_p, transform_gid, transform_s_w):
        a = "ABC"
        b = "DEF"
        c = "GHI"
        z = "something_longer"
        y_a = transform_y_a([0, 1, 1, 1, 0, 1, 1, 1])
        y_p = transform_y_p([0, 1, 1, 1, 1, 0, 0, 1])
        gid = transform_gid([a, z, a, b, b, c, c, c])
        s_w = transform_s_w([1, 1, 1, 5, 5, 7, 7, 7])

        grouped_metric_func = metrics.make_group_metric(mock_func_weight)
        result = grouped_metric_func(y_a, y_p, gid, s_w)
        assert result.overall == 28
        assert len(result.by_group) == 4
        assert result.by_group[a] == 1
        assert result.by_group[b] == 5
        assert result.by_group[c] == 21
        assert result.by_group[z] == 1
        assert result.minimum == 1
        assert result.maximum == 21
        assert result.argmin_set == {a, z}
        assert result.argmax_set == {c}
        assert result.range == 20
        assert result.range_ratio == pytest.approx(1.0/21.0)