Closed
Description
Hi,
Unfortunately, PLSRegression does not work in VotingRegressor. Likely related to PLSRegression returning predictions in shape of (n_samples, 1) instead of (n_samples,) like other single-target regressors.
thanks!
import numpy as np
from sklearn.linear_model import LinearRegression
from sklearn.cross_decomposition import PLSRegression
from sklearn.ensemble import VotingRegressor
r1 = LinearRegression()
r2 = PLSRegression()
X = np.array([[1, 1], [2, 4], [3, 9], [4, 16], [5, 25], [6, 36]])
y = np.array([2, 6, 12, 20, 30, 42])
r1.fit(X, y).predict(X).shape(6,)
r2.fit(X, y).predict(X).shape(6, 1)
er = VotingRegressor([('lr', r1), ('plsr', r2)])
er.fit(X, y).predict(X)Traceback (most recent call last):
Cell In[25], line 1
er.fit(X, y).predict(X)
File ~/code/scikit-learn/sklearn/ensemble/_voting.py:625 in predict
return np.average(self._predict(X), axis=1, weights=self._weights_not_none)
File ~/code/scikit-learn/sklearn/ensemble/_voting.py:69 in _predict
return np.asarray([est.predict(X) for est in self.estimators_]).T
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 2 dimensions. The detected shape was (2, 6) + inhomogeneous part.Metadata
Metadata
Assignees
Labels
No labels