David Holzmüller
David Holzmüller
- I think the old behavior was to produce the original datatype (dataframe or numpy array), but I don't know what we actually want. - Do we want to support...
Interface looks good to me! Did you test what happens if you pass in an all-nan column?
Yes, that is what I would expect. Just wasn't sure if it's going to work with the nanmax/nanmin operations.
Just a thought, having quantile_range as a tuple instead of two separate values might make it harder for people to tune it in some hyperparameter optimization. I think it's probably...
> This can be addressed in another PR you mean, another one for the same release? because it's not as convenient to change the parametes after release, right?
okay, then let's leave it like it is.
Maybe adding an option for the user to specify a batch size would be a good compromise?
I think besides the raw implementation there are two main questions to me that need to be solved: - How to handle categoricals? For a first draft one could ordinal-encode...