[question] Treating high-cardinality categorical features as numeric

[sflender]

Hi LightGBM team,

in the documentation you write that

"For a categorical feature with high cardinality (#category is large), it often works best to treat the feature as numeric [...]"

This is also what I find empirically for cardinalities above 1K-10K. Do you have insights why this is the case? Why does the Fisher algorithm not work well in this case?

[sflender]

resolving as a duplicate of #1934

[github-actions]

This issue has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this.