def test_compute_sample(self, input_data, expected_value):
params = input_data.copy()
vals = {}
vals["y_pred"] = params.pop("y_pred")
vals["y"] = params.pop("y")
metric = ConfusionMatrixMetric(**params)
metric(**vals)
result, _ = metric.aggregate()[0]
np.testing.assert_allclose(result, expected_value, atol=1e-4, rtol=1e-4)
def test_clf_with_nan(self, input_data, expected_value):
params = input_data.copy()
vals = {}
vals["y_pred"] = params.pop("y_pred")
vals["y"] = params.pop("y")
metric = ConfusionMatrixMetric(**params)
result = metric(**vals)
np.testing.assert_allclose(result, expected_value, atol=1e-4, rtol=1e-4)
result, _ = metric.aggregate(reduction="mean_channel")[0]
expected_value, _ = do_metric_reduction(expected_value, "mean_channel")
expected_value = compute_confusion_matrix_metric("tpr", expected_value)
np.testing.assert_allclose(result, expected_value, atol=1e-4, rtol=1e-4)
def test_compute_sample_multiple_metrics(self, input_data, expected_values):
params = input_data.copy()
vals = {}
vals["y_pred"] = params.pop("y_pred")
vals["y"] = params.pop("y")
metric = ConfusionMatrixMetric(**params)
metric(**vals)
results = metric.aggregate()
for idx in range(len(results)):
result = results[idx][0]
expected_value = expected_values[idx]
np.testing.assert_allclose(result, expected_value, atol=1e-4, rtol=1e-4)