csslc / CCSR

Official codes of CCSR: Improving the Stability of Diffusion Models for Content Consistent Super-Resolution
https://csslc.github.io/project-CCSR/
436 stars 32 forks source link

questions about metrics #11

Closed houjie8888 closed 8 months ago

houjie8888 commented 8 months ago

Hi author, thanks for your team's contribution.

I would like to ask you a question about calculating the metrics during the training process. Specifically, the training process is usually interspersed with a validation step, do you perform the computation of the evaluation metrics during the validation step, which seems to be time consuming. So I'm wondering how you schedule the evaluation during the training process?

csslc commented 8 months ago

Hi, thanks for your attention! As you find, performing the inference and the evaluation is time-consuming for diffusion model-based methods, so we did not verify the performance of the model during training. We only tested once or twice during training to verify that our training code was correct.

houjie8888 commented 8 months ago

@csslc Thank you very much for your reply!