allow 'objective' for xgboost to be passed as an engine argument

[topepo]

No description provided.

[jaredlander]

Does this have something to do with the newish warning message in xgboost about the default evaluation metric for binary:logistic being changed from 'error' to 'logloss' as shown below?

WARNING: amalgamation/../src/learner.cc:1061: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.

[juliasilge]

No, I'm pretty sure not. We set the objective but I don't think we do anything with the internal evaluation metric for xgboost.

[jaredlander]

Looking at the fitted model after fitting I see this

call:
  xgboost::xgb.train(params = list(eta = 0.3, max_depth = 4, gamma = 0, 
    colsample_bytree = 1, min_child_weight = 1, subsample = 0.6, 
    objective = "binary:logistic"), data = x$data, nrounds = 100, 
    watchlist = x$watchlist, verbose = 0, nthread = 1)

It looks like you set the watchlist argument: watchlist = x$watchlist

When the watchlist is set I think {xgboost} automatically chooses an eval_metric which in turn causes that warning.

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.