def test_fnr_diff_edge2(self):
# Metric
metric = BinaryFairnessMetrics.FNRDifference()
# edge case of -1
y_true = np.array([0, 1, 1, 0, 1, 1, 1, 0, 1, 0])
y_pred = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])
is_member = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])
assert metric.get_score(y_true, y_pred, is_member) == -1
def test_fnr_diff_normal_invalid(self):
# Metric
metric = BinaryFairnessMetrics.FNRDifference()
# Data
y_true = np.array([0, 1, 1, 0, 1, 1, 1, 0, 1, 2])
y_pred = np.array([0, 0, 1, 0, 0, 1, 1, 1, 0, 0])
is_member = np.array([1, 1, 1, 1, 1, 0, 0, 0, 0, 0])
with self.assertRaises(ValueError):
metric.get_score(y_true, y_pred, is_member)
def test_fnr_diff_normal_list(self):
# Metric
metric = BinaryFairnessMetrics.FNRDifference()
# Data
y_true = [0, 1, 1, 0, 1, 1, 1, 0, 1, 0]
y_pred = [0, 0, 1, 0, 0, 1, 1, 1, 0, 0]
is_member = np.array([1, 1, 1, 1, 1, 0, 0, 0, 0, 0])
assert np.isclose(metric.get_score(y_true, y_pred, is_member),
0.333,
atol=0.001)
def test_fnr_diff_normal_df(self):
my_df = pd.DataFrame.from_dict({
'y_true': [0, 1, 1, 0, 1, 1, 1, 0, 1, 0],
'y_pred': [0, 0, 1, 0, 0, 1, 1, 1, 0, 0],
'is_member': [1, 1, 1, 1, 1, 0, 0, 0, 0, 0]
})
# Metric
metric = BinaryFairnessMetrics.FNRDifference()
# Score
score = metric.get_score(my_df['y_true'], my_df['y_pred'],
my_df['is_member'])
assert np.isclose(score, 0.333, atol=0.001)
