I am handling large data sets (3500 elements) that are too large to expand into discrete columns (MSSLQ has a 1024 column limit), so I am handling them as arrays. I'd like to be able to perform basic ML prep on this data, but it is pretty difficult to do without tossing it straight out to python. Examples of this are normalizing and scaling the data. Would it be possible to identify when an array contains integers or floats and enable some operations in the prep recipe that are specific to this. (in my case, I'm not sure if I want to scale [0,1] or [-1,1], but I'd like to try both) If this is already a feature I've been unable to uncover it through the tool, as well as searching the community forums.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.