LightGBM option MaximumCategoricalSplitPointCount has wrong doc

[rauhs]

See:

machinelearning/src/Microsoft.ML.LightGbm/LightGbmTrainerBase.cs

Lines 189 to 196 in 8b1b14f

	/// <summary>
	/// When the number of categories of one feature is smaller than or equal to <see cref="MaximumCategoricalSplitPointCount"/>,
	/// one-vs-other split algorithm will be used.
	/// </summary>
	[Argument(ArgumentType.AtMostOnce, HelpText = "Max number of categorical thresholds.", ShortName = "maxcat")]
	[TlcModule.Range(Inf = 0, Max = int.MaxValue)]
	[TlcModule.SweepableDiscreteParam("MaxCatThreshold", new object[] { 8, 16, 32, 64 })]
	public int MaximumCategoricalSplitPointCount = 32;

and

machinelearning/src/Microsoft.ML.LightGbm/LightGbmTrainerBase.cs

Lines 49 to 59 in 8b1b14f

	private protected static Dictionary<string, string> NameMapping = new Dictionary<string, string>()
	{
	{nameof(MinimumExampleCountPerLeaf), "min_data_per_leaf"},
	{nameof(NumberOfLeaves), "num_leaves"},
	{nameof(MaximumBinCountPerFeature), "max_bin" },
	{nameof(MinimumExampleCountPerGroup), "min_data_per_group" },
	{nameof(MaximumCategoricalSplitPointCount), "max_cat_threshold" },
	{nameof(CategoricalSmoothing), "cat_smooth" },
	{nameof(L2CategoricalRegularization), "cat_l2" },
	{nameof(HandleMissingValue), "use_missing" }
	};

and
https://lightgbm.readthedocs.io/en/latest/Parameters.html#max_cat_threshold :

max_cat_threshold, default = 32, type = int, constraints: max_cat_threshold > 0
- used for the categorical features
- limit the max threshold points in categorical features

It got mixed up with:

https://lightgbm.readthedocs.io/en/latest/Parameters.html#max_cat_to_onehot

when number of categories of one feature smaller than or equal to max_cat_to_onehot, one-vs-other split algorithm will be used

[najeeb-kazmi]

Interestingly, the helptext has the correct description but we somehow mixed it up in the xml doc