qqingzheng/AI-Self-Training-DPO-SDXL

Issues

• Question on fine-tuning Stable Diffusion

  #5 opened 2 months ago by TomLucidor
  0

• Poor dpo_beta default?

  #4 opened 9 months ago by feffy380
  0

• data question

  #3 opened a year ago by unwritten
  2

• 3 parameters in build_dataset.py but 6 parameters in train.py

  #2 opened a year ago by bronzeman2022
  3

• learning rate adjustment

  #1 opened a year ago by 1073521013
  8