DOC: One-hot vs. Dummy Encoding about formulaic HOT 3 CLOSED

Going by the patsy docs on dummy encoding:
https://patsy.readthedocs.io/en/latest/API-reference.html#categorical-coding-ref
and the scikit-learn docs on one-hot encoding:
https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html#sklearn.preprocessing.OneHotEncoder
I think the difference between dummy encoding and one-hot encoding is whether the person describing the encoding has a statistics background or a machine learning background.

from formulaic.

Good catch. Under the hood the columns are one-hot encoded, but then during the rank reduction steps this transforms most of the time (but not always) to a dummy encoding. I'll keep this issue around so I can hopefully clarify this :).

from formulaic.

Aye... looks like there is no difference after all (but I've added both nomenclatures to ensure that people familiar with the different nomenclature are not confused. Thanks again!

from formulaic.