Added
- add default parameters to preprocess function
- add parameter 'label_col' in preprocess() to remove
rows where the label_col is null
- add parameter 'label_col' in remove_outliers() so that
the label col will not be considered as a column to remove outliers from
- add suggest_col_drop() to suggest which columns to drop (irrelevant columns) based on the column name; like ID, first_name, surname, ...
Changed
- sanitize_column_names() now converts CamelCase to snake_case
- suggest_conversion_dict() now suggests 'category' dtype to numeric columns that are actually categorical; like 'has_credit_card' that has 1 or 0 as values