R bloggers

Tuning xgboost in R: Part I

FavoriteLoadingAdd to favorites

By Gabriel Vasconcelos Before we begin, I would like to thank Anuj for kindly including our blog in his list of the top40 R blogs! Check out the full list at his page, FeedSpot! Tuning a Boosting algorithm for the first time may be a very confusing task. There are so many parameters to choose and they all have different behaviour on the results. Also, the best choice may depends on the data. Every time I get a new dataset I learn something new. A good understanding of classification and regression trees (CART) is also helpful because we will be boosting trees, you can start here if you have no idea of what a CART is. My favourite Boosting package is the xgboost, which will be used in all examples below. Before going to the data let’s talk about some…
Original Post: Tuning xgboost in R: Part I