0

For my project I need to understand and being able to execute this github repository about commonsense generation using the GPT transformer language model. It is quite extensive and I don't have enough programming experience to make sense of it all. Is there anyone who is good with these subjects who can guide me through it/help me?

Or, is there another spot where I can post this question?

Nick ODell
  • 15,465
  • 3
  • 32
  • 66
m.b
  • 45
  • 1
  • 4
  • Are you seeking information on transformer model, or on the general way to train/evaluate a generative model ? You can check out [ai](https://ai.stackexchange.com/help/on-topic) or [cross-validated](https://datascience.stackexchange.com/help/on-topic) stack exchange platforms. – ygorg Apr 12 '21 at 12:30
  • Thanks for your response! Actually the theory about it is clear to me, but their implementation is not, I cannot see the structure in their repository so it is hard for me to replicate. – m.b Apr 12 '21 at 12:44
  • I have a hard time understanding what information you are seeking, and what is your goal. If you want to execute the repository you shared, have you tried following the README ? Did you encounter an error ? Are you trying to use it for your own data and don't know how to do that ? Please add more context to your question. – ygorg Apr 12 '21 at 13:41
  • I am sorry, I will explain. I have read the readme and can also do the experiments. Only I want to replicate the ConceptNet experiment with my own conceptnet train/test set and eventually my own pretrained gpt model. And compare the outcomes. Therefore, I need to understand how this repository is set up, in order for me to be able to change the right things into my own format (different train/dev/test set from conceptnet and my own model). When I want to alter those things I don't know where to start. Is my question more clear? :) – m.b Apr 12 '21 at 14:19

0 Answers0