XGBoost gives warning about not using the parameter max_depth (R)

[shawnbrar]

I had just installed XGBoost on my Ubuntu 18.04. I was trying out the XGBoost R Tutorial. The library was working quiet properly. However, when I was in the ####Verbose Option section of the tutorial, when I would set verbose = 2, my output was not similar to how it was portrayed in the tutorial. It was the same output that I would get if I had put verbose = 1.

Another thing, when I was at the ### Linear Boosting section, if I were to use the following code:-

bst <- xgb.train(data = dtrain, max_depth = 2, eta = 1, nthread = 2, nrou nds = 2, watchlist = watchlist, objective = 'binary:logistic', verbose = 2, booster = 'gblinear')

I would get the following error:-

[19:29:10] WARNING: /home/shawn/Documents/Libs/xgboost/src/learner.cc:485:
Parameters: { max_depth } might not be used.

This may not be accurate due to some parameters are only used in language bindings but
passed down to XGBoost core. Or some parameters are not used but slip through this verification. Please open an issue if you find above cases.

[1] train-error:0.027330 test-error:0.022346
[2] train-error:0.004606 test-error:0.003724

However, if I were to remove the booster argument form the xgb.train function, I do not get the warning about the { max_depth } might not be used

[trivialfis]

Thanks for rasing the issue, linear booster doesn't have depth, only tree has. So it's a correct warning.

[trivialfis]

The doc is incorrect. Reopening.