本来想做三人LORA,结果参数很好LOSS=0.1,然而是个失败品,所以单生成了一个低学习率的(loss=0.078)大姐,意外好用,明后天再把两妹妹LORA上传。
原图是6??X1080,推荐生成图大小400X800以上,精修X1,6倍EXRGANx4会有不错效果。
I originally intended to do a three person LORA, but the parameter was very good, and the loss=0.1. However, it was a failure product, so I generated a low learning rate (loss=0.78) elder sister(dajie), which unexpectedly worked well. I will upload the LORA for my two younger sisters tomorrow and the day after.
The original image is 6??X1080, and it is recommended to generate an image size of 400X800 or above. Refined X1,6 times EXRGANx4 will have a good effect.
Description
目前最好的二姐,第10样本学习LOSS=0.0998,成图由384X768画面放大1.5倍,upscaler R-ESRGAN 4X+到576X1152(能整除??,效果最好)。
At present, the best second sister, the 10th sample learning loss=0.0998, zoomed in by 1.5 times from 384X768, upscaler R-ESRGAN 4X+to 576X1152 (can divide by ??, the best effect).
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.







