I have a spark dataframe with hundreds of columns. Most of the columns are fine as they are, however there are some location columns which are dummied (like binary), for example
