document rounding behavior of floating point numbers in categorical features

docs/Parameters.rst

@@ -828,6 +828,8 @@ Dataset Parameters
 
    -  **Note**: the output cannot be monotonically constrained with respect to a categorical feature
 
+   -  **Note**: floating point numbers in categorical features will be rounded towards 0
+
 -  ``forcedbins_filename`` :raw-html:`<a id="forcedbins_filename" title="Permalink to this parameter" href="#forcedbins_filename">&#x1F517;&#xFE0E;</a>`, default = ``""``, type = string
 
    -  path to a ``.json`` file that specifies bin upper bounds for some or all features

include/LightGBM/config.h

@@ -703,6 +703,7 @@ struct Config {
   // desc = **Note**: using large values could be memory consuming. Tree decision rule works best when categorical features are presented by consecutive integers starting from zero
   // desc = **Note**: all negative values will be treated as **missing values**
   // desc = **Note**: the output cannot be monotonically constrained with respect to a categorical feature
+  // desc = **Note**: floating point numbers in categorical features will be rounded towards 0
   std::string categorical_feature = "";
 
   // desc = path to a ``.json`` file that specifies bin upper bounds for some or all features