Four score and rule 34 ago.....
If you like my contributions to the community, please consider Buying Me A Coffee or check out my other sites on AllMyLinks
Trained on 14 photos
1596 steps
1 epoch
Trained at 768 resolution, so 768x768 inpainting works great
For txt2img, a weight of 0.7-0.9 will give a little more flexibility, but 1.0 is noise-free. A weight of 1.0 works really well for inpainting. Face restoration isn't usually necessary, but I recommend setting your face restoration strength inside auto1111 settings to something like 0.7-0.8 which makes the effect less than the default setting of 0.5.
The checkpoint you use can heavily influence the likeness of a celebrity Lora. I train my Lora's on SD15, so you'll get the best likeness on any model that used SD15 as the base model. See this example for how a checkpoint can influence the look of a Lora.