[GitHub] madlib pull request #301: DT/RF: Don't eliminate single-level categorical va...

[jingyimei]

Github user jingyimei commented on a diff in the pull request:

    https://github.com/apache/madlib/pull/301#discussion_r205910952
  
    --- Diff: 
src/ports/postgres/modules/recursive_partitioning/test/decision_tree.sql_in ---
    @@ -282,13 +283,16 @@ SELECT tree_train('dt_golf'::text,         -- source 
table
                              6::integer,        -- min split
                              2::integer,        -- min bucket
                              3::integer,        -- number of bins per 
continuous variable
    -                         'cp=0.01, n_folds=2'          -- cost-complexity 
pruning parameter
    +                         'cp=0.01, n_folds=2',          -- cost-complexity 
pruning parameter
    +                         'null_as_category=True'
                              );
     
     SELECT _print_decision_tree(tree) from train_output;
     SELECT tree_display('train_output', False);
     SELECT impurity_var_importance FROM train_output;
    +SELECT cat_levels_in_text, cat_n_levels, impurity_var_importance FROM 
train_output;
    --- End diff --
    
    Is it better to assert cat_n_levels == {1} instead of just quering?


---