I want to post prune my decision tree as it is overfitting, I can do this using cost complexity pruning by adjusting ccp_alphas parameters however this does not seem very intuitive to me.

From my understanding there are some hyperparameters such as min_samples_split, max_depth, min_impurity_split, min_impurity_decrease that will prune my tree to reduce overfitting.

Since I am working with a larger dataset it takes a long time to train therefore don't want to just do trial-error.

What are some possible combinations of above mentioned hyperparameters that will prune my tree? + reasoning behind choosing particular combination will be helpful.

Thanks in advance!


1 Answer 1


There are no combinations that work for all cases, hyperparameter tuning is still something that is mostly done by trial and error. Things like Gridsearch and Randomsearch exist though.

A good start is always the default setting. An idea if performance is an issue is to tune on a small percentage of the training set to later switch to the full set.


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.