def test_pred_equality_edge_5(self):
# edge case of - 1
y_pred = np.array([0, 0, 0, 0, 1, 1, 1, 1])
y_true = np.array([0, 0, 0, 1, 1, 1, 1, 0])
is_member = np.array([1, 1, 1, 1, 0, 0, 0, 0])
metric = BinaryFairnessMetrics.PredictiveEquality()
assert metric.get_score(y_true, y_pred, is_member) == -1
def test_pred_equality_edge_4(self):
# Data: edge case - homogeneous ground truth within group - returns None
# edge case of 1
y_true = np.array([0, 0, 0, 1, 1, 1, 1, 0])
y_pred = np.array([1, 1, 1, 1, 0, 0, 0, 0])
is_member = np.array([1, 1, 1, 1, 0, 0, 0, 0])
# Metric
metric = BinaryFairnessMetrics.PredictiveEquality()
# Score
assert metric.get_score(y_true, y_pred, is_member) == 1
def test_pred_equality_edge_1(self):
# Data: edge case - homogeneous ground truth within group - returns None
# unprivileged homogeneous
y_true = np.array([0, 0, 0, 1, 1, 1, 1, 1])
y_pred = np.array([0, 0, 0, 0, 0, 0, 0, 0])
is_member = np.array([1, 1, 1, 1, 0, 0, 0, 0])
# Metric
metric = BinaryFairnessMetrics.PredictiveEquality()
with self.assertWarns(UserWarning):
assert metric.get_score(y_true, y_pred, is_member) is None
def test_pred_equality_normal_invalid(self):
# Data: edge case - homogeneous ground truth within group - returns None
# medium number
y_true = np.array([0, 1, 0, 1, 1, 1, 1])
y_pred = np.array([0, 0, 1, 0, 0, 1, 1, 1])
is_member = np.array([1, 1, 1, 1, 0, 0, 0, 0])
# Metric
metric = BinaryFairnessMetrics.PredictiveEquality()
# Score
with self.assertRaises(InputShapeError):
metric.get_score(y_true, y_pred, is_member)
def test_pred_equality_normal_list(self):
# Data: edge case - homogeneous ground truth within group - returns None
# medium number
y_true = [0, 1, 0, 1, 1, 1, 1, 0]
y_pred = [0, 0, 1, 0, 0, 1, 1, 1]
is_member = [1, 1, 1, 1, 0, 0, 0, 0]
# Metric
metric = BinaryFairnessMetrics.PredictiveEquality()
# Score
score = metric.get_score(y_true, y_pred, is_member)
assert np.isclose(score, -0.5, atol=0.001)
def test_pred_equality_normal_df(self):
# medium number
my_df = pd.DataFrame.from_dict({
'y_true': [0, 1, 0, 1, 1, 1, 1, 0],
'y_pred': [0, 0, 1, 0, 0, 1, 1, 1],
'is_member': [1, 1, 1, 1, 0, 0, 0, 0]
})
# Metric
metric = BinaryFairnessMetrics.PredictiveEquality()
# Score
score = metric.get_score(my_df['y_true'], my_df['y_pred'],
my_df['is_member'])
assert np.isclose(score, -0.5, atol=0.001)
