Is there an easy way to output the tokenized (xxmaj, xxup, xxbos, etc.) text to a csv file? All of my raw data is in a csv and the fastai library is working great, but I want to compare with another model (as a baseline) and want to input the exact same tokenized text that is sent to the ULMFit (language_learner) model.
Thanks for your help,
Esteban