search

dmlc / xgboost

Scalable, Portable and Distributed Gradient Boosting (GBDT, GBRT or GBM) Library, for Python, R, Java, Scala, C++ and more. Runs on single machine, Hadoop, Spark, Dask, Flink and DataFlow
https://xgboost.readthedocs.io/en/stable/
Apache License 2.0
26.3k stars 8.73k forks source link

min_child_weight is not used for Tweedie objective function #9324

Closed yellowzijian closed 1 year ago

yellowzijian commented 1 year ago

Description: When using R to perform Tweedie regression, I have encountered an issue regarding the min_child_weight parameter. After specifying this parameter in the list of parameters passed to the xgb.train or xgboost function, it seems that the value provided for min_child_weight is not utilized during the training process. This behavior appears to be specific to the R implementation, as I believe the parameter is successfully utilized in a similar procedure using Python.

WARNING from R: [10:14:45] WARNING: src/learner.cc:767: Parameters: { "mini_child_weight" } are not used.

houlk8503 commented 1 year ago

The warning raises because you have a spelling error. The correct parameter name should be 'min_child_weight', but what you write down the in R script, as reflected in your warning message, is 'mini_child_weight' (it starts with 'mini', not 'min').

trivialfis commented 1 year ago

@houlk8503 Thank you for pointing that out.