How can I specify the number of the trees in my xgboost model ,using xgb.train?

Question

If I use xgb.train, how can i specify the number of the trees in my model ?

In xgboost.XGBRegressor(), I know i can use parameter 'n_estimators'， but what should I do in xgb.train()?

I searched Google and didn't find any answer, thanks in advance.

Answer 1

Its num_boost_round. As stated on the api documentation:

num_boost_round (int) – Number of boosting iterations.

To exactly match the output of xgb.train and scikit wrapper (XGBRegressor or XGBClassifier) you can look at my other answer here:

• https://stackoverflow.com/a/46947191/3374996