I have a gbm problem and I am using the gbm package in R for it. But in most forums I see people using caret package for gbm. Is there any advantage of using caret instead of gbm package? If so, what sort of advantages do we have, and why?
See here: http://www.jstatsoft.org/article/view/v028i05
caret can be seen (among other relevant advantages, to be sure) as a "wrapper" for packages like
gbm, so in terms of statistical results they should produce identical results (given identical tuning parameters, of course). So ease of use may favor
caret, especially if you aim to use other ML techniques at the same time.