为什么用你们的框架训练出的stable diffusion ckpt 推理结果异常为噪声图 #2460
Unanswered
yufengyao-lingoace
asked this question in
Community | Q&A
Replies: 1 comment 3 replies
-
因为,因为diffusion模型本身训练需要很长时间,如果train from scratch的话,训练时间比较短就是没有什么效果。 |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
为什么用你们的框架训练出的stable diffusion ckpt 推理结果为噪声图,训练过程看着正常,loss也在逐渐收敛,但是txt2img.py推理出的图异常如下(完全按照你们官方文档复现):
会不是是框架存在问题,导致保存的权重有问题。
你们那边有没有好的建议?
Beta Was this translation helpful? Give feedback.
All reactions