Is there any way to use sklearn on GPU?

Question

I am training model using a VotingClassifier in sklearn. The dataset is large, approximately 1 millions tabular rows.

from sklearn.ensemble import RandomForestClassifier
from sklearn.ensemble import VotingClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.svm import SVC

log_clf = LogisticRegression(max_iter=10000)
rnd_clf = RandomForestClassifier(n_estimators=50)
svm_clf = SVC()
voting_clf = VotingClassifier(
                estimators=[('lr', log_clf), ('rf', rnd_clf), ('svc', svm_clf)],
                voting='hard'
            )

I did some research and there are cuML available for sklearn to work on GPU. But cuml doesn't support VotingClassifier yet. Is there any other ways to train sklearn model on GPU at the moment?

from cuml.ensemble import RandomForestClassifier as cuRFC
from sklearn.ensemble import VotingClassifier
from cuml.linear_model import LogisticRegression
from cuml.svm import LinearSVC

I tried to use voting classifier with cuML models. But it doesn't work.

Answer 1

By default it does not use GPU, especially if it is running inside Docker, unless you use nvidia-docker and an image with a built-in support. Scikit-learn is not intended to be used as a deep-learning framework and it does not provide any GPU support.