xuanyicode

I am not quite sure how big an influence the problem has on the neural network training process. Is it possible to skip this problem?

For reference, I once meet similar problem, that is because I set one layer of the model to be loaded on CUDA device independently, when I remove the loading, it...

thanks for your helpful code, and I have an extended question: how to input several images in a multi-turn dialogue?

same question!