Each reader has its own preprocessing configuration and type guessing, thus each reader recognises feature roles differently.
Table below summarizes the guessed feature roles for all applied presets.
"N" - numerical, "C" - categorical, "D" - datetime, "T" - textual, "-" - dropped.
{{ roles_table }}Below is the precise description of training data for each preset.
{% for section in data_sections %} {{ section }} {% endfor %}