I'm new to the machine learning field, and I'm learning ML models by practice, and I'm facing an issue while using the machine learning model.

While I'm implementing the RandomForestClassifier model with hyper tunning it's taking too much time to predict output. And I'm also using GridSearchCV on it. so it's take much time.

Is there any way how can I solve this problem.

OR, Is Google Colab or Kaggle Notebook editor can perform better than Jupiter Notebook ?

  • 1
    $\begingroup$ Are you training with a GPU? If no then it definitely will be slow. Yes GridSearchCV is very slow when it comes to hyperparameter optimization even when training with a GPU. You could use RandomSearchCV which is faster but the best option would be to use a Bayesian Optimizer. A library I would recommend for this is Hyperopt. $\endgroup$
    – yudhiesh
    Commented Sep 19, 2020 at 14:10
  • $\begingroup$ Okay, My computer doesn't have any GPU, so is there any way to run it on Google Colab or Kaggle Kernal? Or it provides GPU facility. $\endgroup$ Commented Sep 20, 2020 at 8:41
  • $\begingroup$ I will add an answer $\endgroup$
    – yudhiesh
    Commented Sep 20, 2020 at 8:46

1 Answer 1


You can access the GPU by going to the settings:

Runtime> Change runtime type and select GPU as Hardware accelerator.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.