CODEWITHSUNDEEP
pythonmachine-learningscikit-learn
dwitvliet
dwitvliet

Reputation: 7671

Get the score of a single sample in a multivariable model?

I'm trying to make a simple scikit-learn example work, but I keep getting the error: multiclass-multioutput is not supported.

The first part of my code, which follows any basic tutorial, works as predicted:

>>> from sklearn import ensemble
>>> from sklearn import datasets
>>> dataset = datasets.load_linnerud()
>>> X, y = dataset.data, dataset.target
>>> clf = ensemble.RandomForestClassifier(n_estimators=500)
>>> clf.fit(X, y)
>>> clf.predict([X[0]])
array([[ 191.,   36.,   50.]])

However, when I try to get a prediction score, I get the following error:

>>> clf.score([X[0]], [y[0]])
(...)
ValueError: multiclass-multioutput is not supported

When working with datasets.load_iris(), everything works fine. What am I doing wrong? How can I get the score of a single sample in a multivariable model?

Upvotes: 0

Views: 239

Answers (1)

Kyle Kastner
Kyle Kastner

Reputation: 1018

Try swapping RandomForestRegressor in for RandomForestClassifier. The docs for load_linnerud() show that it is a regression problem. Doing that made the code work for me.

Upvotes: 2

Related Questions