Is there a way I can create a new column using pandas 'pd.cut' and 'pd.qcut' functions in visidata on a specific column?

[oostopitre]

Is there a way I can create a new column in visidata using pandas 'pd.cut' and 'pd.qcut' functions on an existing numeric column?

I tried importing pandas in my .visidatarc but I am not able to figure out how to use these specific pandas functions on a column. I tried searching similar threads but wasn't able to figure out quite how to accomplish this. I think this might be because to use these functions, it needs info on the values of the whole column.

Any tips or help is super appreciated. Love visidata!

[midichef]

Hi, this should work:

In your .visidatarc, put import pandas as pd
Make a Python list holding the current column's values:
Space exec-python vals = [vd.sheet.cursorCol.getTypedValue(r) for r in sheet.rows]
Process it with Pandas cut():
Space exec-python cutvals = pd.cut(vals, bins=[0,3,6,9], labels=['one','two','three'])
Then make a new column whose rows are the elements of cutvals:
za col_cuts
gs gz= cutvals

If you get an error like TypeError: '<' not supported between instances of 'int' and 'str', the problem may be that your starting column was a string. Make sure it has a numeric type, by pressing # or % or $.

If you want to run this command often, make a macro, and generalize the cut command to refer to your own custom variables, e.g. cutbins and cutlabels:
Space exec-python cutvals = pd.cut(vals, bins=cutbins, labels=cutlabels)

It may be possible in the future to make a new column more simply, with some changes to aggregators that are still being considered.

[oostopitre]

Thank you for the suggestion. Will try this out. This will give me atleast a temporary way to do this.