How to check if sklearn model is classifier or regressor

Question

Is there a simple way to check if a model instance solves a classification or regression task in the scikit-learn library?

This is a funny thing to want, can't really imagine a scenario in which that info is not available prior to actually fitting a model. I can only think of checking what kind of data the model is predicting — yatu, Oct 01 '19 at 13:26

score 20 · Accepted Answer · answered Oct 02 '19 at 07:53

Use sklearn.base.is_classifier and/or is_regressor:

from sklearn.base import is_classifier, is_regressor
from sklearn.linear_model import LinearRegression
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import RandomForestClassifier

models = [LinearRegression(), RandomForestClassifier(), RandomForestRegressor()]

for m in models:
    print(m.__class__.__name__, is_classifier(m), is_regressor(m))

Output:

# model_name is_classifier is_regressor
LinearRegression False True
RandomForestClassifier True False
RandomForestRegressor False True

Thanks, this is exactly what I was looking for! I didn't know it existed in the base package. — the_man_in_black, Oct 02 '19 at 09:06

score 1 · Answer 2 · answered Oct 01 '19 at 13:46

I guess you ask this because you have a serialized model whose type you do not know. Open the file and do

mlType = type(variable_name)

where variable_name is the handle of your de-serialized model.

output e.g.

class 'sklearn.linear_model.base.LinearRegression'

How to check if sklearn model is classifier or regressor

2 Answers2