How to fine-tune G_mix on my own style dataset?

[yuepengfeidev]

Thank you for your excellent work.
I would like to use my own style dataset for image generation. To achieve better results, I want to fine-tune G_mix on my style dataset. I have already built the .lst, .rec, and .idx files for my style dataset. However, I have encountered some issues during the training process:

1. Can you provide an overview of the own style dataset construction and training process on own dataset?
2. Do I need to build center.pth and simlist.pth for training? How should I construct them?

Thank you very much. I am looking forward to your response.