What is the pros and cons of using XGBoost VS GBR (scikit-learn) when dealing with data 500<records<1000 and about 5 columns?
1 Answer
$\begingroup$ $\endgroup$
XgGboost is a more complex and powerful model compared to the GBR model The size of your dataset is small and XGBoost might get overfitted on the dataset and the dataset is small and too simple for it
GBR doesn't have thr missing value handling feature But for such a small dataset one can easily use either method to fill them missing completely at random(MCAR) , missing at random (MAR), missing not a random (MNAR)
https://stefvanbuuren.name/fimd/sec-MCAR.html
https://www.kdnuggets.com/wtf-is-the-difference-between-gbm-and-xgboost
