Hello,
I have successfully run the prediction of the aurora-0.1-finetuned model on an A100 80G GPU. Next, I hope to continue fine-tuning the aurora-0.1-finetuned model based on my own dataset. I would like to know: regardless of training time, what are the minimum hardware requirements for continuing the fine-tuning? And will a server with 8x A100 80G GPUs be sufficient?
Thank you!!