waterhorse1/ChessGPT

Number of data need for finetunning

khoaguye opened this issue · 0 comments

Hello,
This is a great paper and I am happy that I found it, I am not sure if I missed this in the paper or not, so there are a couple of points that I would appreciate if you could clarify:

  1. I noticed that the accuracy of the move is evaluated based on the ELO rating. Could you please explain how this is calculated?
  2. Could you specify the size of the dataset used to achieve on different score?
    Thank you for your time and assistance.