使用parsnip::set_engine("xgboost")传参触发警告，如何消除？

问题背景 #

使用R语言的parsnip包训练xgboost模型时，尝试通过parsnip::set_engine()传递额外参数（eval_metric、seed），先后尝试两种方式均触发警告：

尝试的两种传参方式 #

方式一：

model_list$xgboost <- parsnip::boost_tree(
      trees = 1000,
      tree_depth = hardhat::tune(),
      learn_rate = hardhat::tune(),
      loss_reduction = hardhat::tune(),
      sample_size = hardhat::tune(),
      mtry = hardhat::tune()
    ) %>%
 parsnip::set_engine("xgboost", 
                    eval_metric = "rmse", 
                    seed = 55
                    ) %>%
 parsnip::set_mode("regression")

方式二：

model_list$xgboost <- parsnip::boost_tree(
      trees = 1000,
      tree_depth = hardhat::tune(),
      learn_rate = hardhat::tune(),
      loss_reduction = hardhat::tune(),
      sample_size = hardhat::tune(),
      mtry = hardhat::tune()
    ) %>%
 parsnip::set_engine("xgboost", 
                     params = list(eval_metric = "rmse", 
                                   seed = 55)
                    ) %>%
 parsnip::set_mode("regression")

触发的警告信息 #

warning: Passed invalid argument 'info' - entries on it should be passed as direct arguments. This warning will become an error in a future version., Passed invalid function arguments: eval_metric, seed, nthread. These should be passed as a list to argument 'params'. Conversion from argument to 'params' entry will be done automatically, but this behavior will become an error in a future version., Parameter 'watchlist' has been renamed to 'evals'. This warning will become an error in a future version., Argument 'objective' is only for custom objectives. For built-in objectives, pass the objective under 'params'. This warning will become an error in a future version.

当前环境：R 4.3.3、parsnip 1.3.2、xgboost 3.0.2.1

解决方法 #

结合xgboost 3.x版本的API变化以及parsnip的封装规则，调整参数传递方式即可消除警告：

修改后的代码 #

model_list$xgboost <- parsnip::boost_tree(
  trees = 1000,
  tree_depth = hardhat::tune(),
  learn_rate = hardhat::tune(),
  loss_reduction = hardhat::tune(),
  sample_size = hardhat::tune(),
  mtry = hardhat::tune()
) %>%
  parsnip::set_engine(
    "xgboost",
    params = list(
      eval_metric = "rmse",
      seed = 55,
      # 回归任务明确指定内置目标函数，避免额外警告
      objective = "reg:squarederror"
    )
  ) %>%
  parsnip::set_mode("regression")

关键注意事项 #

1. 参数分类传递：boost_tree()中已经封装的参数（如trees、tree_depth）直接在该函数中设置，xgboost原生参数（eval_metric、seed、内置objective）必须放入params列表中传递给set_engine()。
2. 适配xgboost 3.x API：
   • 原watchlist参数需替换为evals，同样放入params列表。
   • 内置目标函数不能单独作为set_engine()的参数，必须放在params里，否则会触发objective相关警告。
3. 提前适配未来版本：警告提示当前的兼容转换会在未来版本移除，因此按要求使用params列表传递参数是长期可行的方案。

内容的提问来源于stack exchange，提问作者SiriO