i would like to feed part of speech of each word as input of additional information in BERT model.
for that i would like to concatenate POS as one hot vector along with word embedding.
i would like to know is it possible? if yes please suggest me how to do
thanks