0

I am currently playing on ClipDrop.co with Stable Diffusion, and I am confused with the tutorials I can find online.

For instance, for mixing two celebrities in one pics, I read all the following syntaxes:

  • [Person A:Person B:0.5], where 0.5 is the multiplier to determine the number of steps before switching to Person B.
  • (Person A:Person B)
  • (Person A|Person B)

I've tried multiple ratio, switching A and B, and played with all these syntaxes, but I have no way of really quantifying the differences I can notice in terms of distance to Person A and Person B of the result.

So I start to wonder if the syntax matters and where to find an official manual.

Myoch
  • 101
  • I went through some source code for one of the servers earlier this year and it almost looked that it was turning it into an array an encoding each word before doing inference. So I think a lot of the syntax is just sugar for turning it into an array. – Emil Aug 31 '23 at 06:06
  • The specific one I used seemed to want something like (person:1.2) (that is it wanted floats, not ints). – Emil Aug 31 '23 at 06:37

0 Answers0