def test_negative_with_log():
y_predicted = np.array([-1, 10, 30])
y_true = np.array([-1, 0, 1])
for objective in [MeanSquaredLogError(), RootMeanSquaredLogError()]:
with pytest.raises(
ValueError,
match=
"Mean Squared Logarithmic Error cannot be used when targets contain negative values."
):
objective.score(y_true, y_predicted)
def test_log_linear_model():
obj = MeanSquaredLogError()
root_obj = RootMeanSquaredLogError()
s1_predicted = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2])
s1_actual = np.array([0, 0, 0, 0, 0, 0, 0, 0, 0])
s2_predicted = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2])
s2_actual = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2])
s3_predicted = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2])
s3_actual = np.array([2, 2, 2, 0, 0, 0, 1, 1, 1])
assert obj.score(s1_predicted, s1_actual) == pytest.approx(0.562467324910)
assert obj.score(s2_predicted, s2_actual) == pytest.approx(0)
assert obj.score(s3_predicted,
s3_actual) == pytest.approx(0.617267976207983)
assert root_obj.score(s1_predicted,
s1_actual) == pytest.approx(np.sqrt(0.562467324910))
assert root_obj.score(s2_predicted, s2_actual) == pytest.approx(0)
assert root_obj.score(s3_predicted, s3_actual) == pytest.approx(
np.sqrt(0.617267976207983))
def test_log_metrics_only_passed_directly(X_y_regression):
X, y = X_y_regression
with pytest.raises(
ObjectiveNotFoundError,
match="RootMeanSquaredLogError is not a valid Objective!"):
AutoMLSearch(X_train=X,
y_train=y,
problem_type='regression',
additional_objectives=[
'RootMeanSquaredLogError', 'MeanSquaredLogError'
])
ar = AutoMLSearch(X_train=X,
y_train=y,
problem_type='regression',
additional_objectives=[
RootMeanSquaredLogError(),
MeanSquaredLogError()
])
assert ar.additional_objectives[0].name == 'Root Mean Squared Log Error'
assert ar.additional_objectives[1].name == 'Mean Squared Log Error'
y = pd.Series([2, 3, 0, 1, 1])
invalid_targets_check = InvalidTargetDataCheck(problem_type="regression", objective=objective)
assert invalid_targets_check.validate(X, y) == {
"warnings": [],
"errors": [DataCheckError(
message=f"Target has non-positive values which is not supported for {objective}",
data_check_name=invalid_targets_data_check_name,
message_code=DataCheckMessageCode.TARGET_INCOMPATIBLE_OBJECTIVE,
details={"Count of offending values": sum(val <= 0 for val in y.values.flatten())}).to_dict()],
"actions": []
}
@pytest.mark.parametrize("objective", [RootMeanSquaredLogError(), MeanSquaredLogError(), MAPE()])
def test_invalid_target_data_check_invalid_labels_for_nonnegative_objective_instances(objective):
X = pd.DataFrame({'column_one': [100, 200, 100, 200, 200, 100, 200, 100] * 25})
y = pd.Series([2, 2, 3, 3, -1, -1, 1, 1] * 25)
data_checks = DataChecks([InvalidTargetDataCheck], {"InvalidTargetDataCheck": {"problem_type": "multiclass",
"objective": objective}})
assert data_checks.validate(X, y) == {
"warnings": [],
"errors": [DataCheckError(
message=f"Target has non-positive values which is not supported for {objective.name}",
data_check_name=invalid_targets_data_check_name,
message_code=DataCheckMessageCode.TARGET_INCOMPATIBLE_OBJECTIVE,
details={"Count of offending values": sum(val <= 0 for val in y.values.flatten())}).to_dict()],
"actions": []
message=
f"Target has non-positive values which is not supported for {objective}",
data_check_name=invalid_targets_data_check_name,
message_code=DataCheckMessageCode.
TARGET_INCOMPATIBLE_OBJECTIVE,
details={
"Count of offending values":
sum(val <= 0 for val in y.values.flatten())
}).to_dict()
]
}
@pytest.mark.parametrize(
"objective", [RootMeanSquaredLogError(),
MeanSquaredLogError(),
MAPE()])
def test_invalid_target_data_check_invalid_labels_for_nonnegative_objective_instances(
objective):
X = pd.DataFrame(
{'column_one': [100, 200, 100, 200, 200, 100, 200, 100] * 25})
y = pd.Series([2, 2, 3, 3, -1, -1, 1, 1] * 25)
data_checks = DataChecks(
[InvalidTargetDataCheck], {
"InvalidTargetDataCheck": {
"problem_type": "multiclass",
"objective": objective
}
})