Catch two types of warning

Question

So I want to catch different types of warnings on the following code:

score = cross_val_score(DecisionTreeClassifier(), X, y, scoring=f2_measure, cv=cv)

From running the code, I got two warnings:

UserWarning: The least populated class in y has only 1 members, which is less than n_splits=2.
  % (min_groups, self.n_splits)), UserWarning)

UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 due to no true nor predicted samples. 
  Use `zero_division` parameter to control this behavior.

I'd like to catch both type of warnings and then handle them separately, I tried using the following:

with warnings.catch_warnings():
    warnings.filterwarnings('error', category=UserWarning)
    warnings.filterwarnings('error', category=UndefinedMetricWarning)
    try:
        score = cross_val_score(DecisionTreeClassifier(), X, y, scoring=f2_measure, cv=cv).mean()
    except (UndefinedMetricWarning, UserWarning) as e:
        print(type(e))

But all of the warnings are then raised as a UserWarning, and there's no UndefinedMetricWarning. (I'm not quite sure but I think this problem might be caused by UndefinedMetricWarning is inherited from UserWarning).

How to identify and handle these two type of warnings separately?

Catch two types of warning

0 Answers