Request to add parameters - monotone_constraints, top_rate, other_rate, max_cat_threshold, max_cat_to_onehot, max_conflict_rate, is_enable_sparse, sparse_threshold, num_threads, data_random_seed, feature_fraction_seed, early_stopping_round, use_missing, zero_as_missing, min_data_in_bin, min_data_per_group, min_data_in_leaf, min_gain_to_split, cat_l2, cat_smooth, feature_fraction_bynode

we use these parameter in LightGBM on our non-spark model. We are implementing the same on spark and these parameter are useful in Credit Risk modeling. It would be great if these parameters are added.

monotone_constraints,
top_rate,
other_rate,
max_cat_threshold,
max_cat_to_onehot,
max_conflict_rate,
is_enable_sparse,
sparse_threshold,
num_threads,
data_random_seed,
feature_fraction_seed,
early_stopping_round,
use_missing,
zero_as_missing,
min_data_in_bin,
min_data_per_group,
min_data_in_leaf,
min_gain_to_split,
cat_l2, cat_smooth,
feature_fraction_bynode

Thanks!

The issue names no files, tests, or entry points. Start by locating the Spark LightGBM parameter bindings and comparing them with the listed non-Spark parameters; done means the requested parameters are supported for the Spark implementation.

Written by the indexing model from the issue text.