关于sd-turbo训练问题

cswry / SeeSR

[CVPR2024] SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution

Apache License 2.0

390 stars 25 forks source link

Open Yu-xi-99 opened 4 months ago

Yu-xi-99 commented 4 months ago

Hi，想请教一下基于sd-turbo的训练具体是怎么做的呢？我尝试过仅把sd2-base换成turbo，发现训出来的结果相比baseline要更模糊一些

cswry commented 4 months ago

你好，请问turbo和sd2-base分别跑了多少步呀？

Yu-xi-99 commented 3 months ago

sd2-base推断时是20步，turbo尝试了4步和20步，都很糊