Used deepset/roberta-base-squad2, deepset/tinyroberta-squad2, gpt2, distilgpt2 already trained models and hugging face API to generate questions and tokenize the PDFs uploaded in it.