[dask] test that Dask automatically treats 'category' columns as categorical features

[jameslamb]

Based on thread starting at #3930 (comment).

#3908 added some tests that the lightgbm.dask estimators correctly handled Dask DataFrames where some columns are pandas "category" columns. That PR used the parameter categorical_feature to explicitly provide a list of which features should be treated as categorical.

This is unnecessary, since auto is the default behavior and auto tells LightGBM "treat 'category' columns as categorical features" (

LightGBM/python-package/lightgbm/basic.py

Lines 513 to 514 in 846b512

	if categorical_feature == 'auto': # use cat cols from DataFrame
	categorical_feature = cat_cols_not_ordered

).

This PR removes the unnecessary categorical_features parameter in tests. As of this PR, we'll now be testing not only that lightgbm.dask estimators correctly train on categorical features, but also that it also automatically

[StrikerRUS]

Nice!

[jameslamb]

Nice!

thanks for pointing it out!

[github-actions]

This pull request 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.