Encode_plus Pretokenized input seuqence must be Union

I have a problem with arguments types for the encode_plus method.

The code works when we put, for example, df[‘labels’][0] or any other index (so when it is list of strings), but not with the whole column (df[‘labels’].to_list(), so when it is list of lists of strings).

We need to tokenize the whole data frame, so our question is - is there another function that can do that, or we need to iterate through the data frame and call encode_plus() on each separate cell value?