Currently yes, you will need to refer to each column by name. This is functionality we are looking at improving by enable passing entire dataframes in and out of the product with one variable.
Please feel free to reach out to me at firstname.lastname@example.org if you have more questions or ideas!
Thanks for the reply.
Then my question would be if I'm using some regression in sklearn which always requires input X and y as a numpy array. In tabpy, do i have to convert it to numpy array before putting into algorithm?
For example, my _arg1, _arg2 are X, _arg 3 is y
X = np.array([_arg1, _arg2])
y = np.array(_arg3)
Or in other words, what's the python data type here for a column, is it a list?
Yes, Tableau imports data into Python as lists, and most sklearn models will expect data to be in the form of a numpy array.