How to fine tune a model to give ratings? - GPT builders - OpenAI ...
This will be my first attempt at fine-tuning a model. I aim to develop a model that can provide ratings on the level of fraudulence in an ecommerce product, or simply determine whether it is fraudulent with a 'yes' or 'no' answer. The main questions I have are whether this idea is feasible and, if so, how can I go about achieving it?
To start off, I need to consider whether the concept is practical. Subsequently, I should think about the process of implementing it. Should I provide the model with data on the products I come across, along with the corresponding ratings indicating their legitimacy?
Powered by Discourse, best viewed with JavaScript enabled
