Does a <MISSING> score field indicate that the column was excluded from training?

A  score field typically means that the column was not used during training. This could be due to several reasons—for example, Kumo may have assigned the column with a Null encoder internally, despite being designated as a multicategorical stype. Also, Kumo supports a max of limit of 500 distinct values, so surpassing this threshold would default to a null encoder.