Is there support for multi-label data in the tidymodels ecosystem? I don't seem to be able to specify the model formula to include the various columns that we need to train/predict. Some classifiers seem ok with a string but most seem to want one hot encode.
For context my problem is similar to the classic IMDb film category problem. I have a block of text which can have a single or multiple labels (aka description of Harry Potter can be classified as fantasy, children's etc).
Many thanks
