1

I train GeneticSelectionCV model on cpu with the following code:

from genetic_selection import GeneticSelectionCV
from sklearn.naive_bayes import GaussianNB
g = GeneticSelectionCV(estimator=GaussianNB(), scoring='accuracy').fit(X, y)

How can I train GaussianNB and GeneticSelectionCV on GPU?

desertnaut
  • 57,590
  • 26
  • 140
  • 166
Ofir
  • 590
  • 9
  • 19

1 Answers1

0

I'm not aware of whether this has been tested, but you can likely use cuML's GPU-accelerated GaussianNB with GeneticSelectionCV as a drop-in replacement for scikit-learn's GaussianNB. This recent blog provides some information about potential performance.

Nick Becker
  • 4,059
  • 13
  • 19
  • Thanks, I will try this. What about GeneticSelectionCV on GPU? – Ofir Aug 05 '22 at 05:03
  • I'm not familiar with the library this functionality comes from, but if that's something you'd be looking for you may want to file a feature request on that library's Github repository (if it's open-source and on Github). – Nick Becker Aug 05 '22 at 13:47