GPT-2 requires quite a large amount of processing power in order to be fast. I don't have a HUGE amount of processing power, and it seems a bit of overkill for what I want to do, which is get it to generate 1 or 2 small sentences based on some small texts I train it on.

Is there another, less powerful AI that I can use in Python which might suit these needs? It might not be as great as GPT-2, but in my use case, speed is more valuable.

Thanks in advance.

  • IMHO, the old way to do it was RNN – Thomas Weller Jan 9 at 11:06
  • The old old way to do it is Markov chains, which are pretty simple to implement and if you google you can find various Python implemenations (e.g., here). Have you looked at such approaches? – BrenBarn Jan 10 at 5:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.