Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Dynamic Prompts inside Wildcard files #3

Closed
Dsmasha opened this issue Oct 15, 2022 · 5 comments
Closed

Dynamic Prompts inside Wildcard files #3

Dsmasha opened this issue Oct 15, 2022 · 5 comments

Comments

@Dsmasha
Copy link

Dsmasha commented Oct 15, 2022

Would it be possible to make it so that dynamic prompts are read correctly from inside of the wildcard .txts? For example, I have a wildcard file I use to pick random outfits and would like to attach specific randomly chosen backgrounds to each outfit inside the .txt, but right now the images generate a combination of all the tags inside of { | | | } markers rather than selecting one.

@adieyal
Copy link
Owner

adieyal commented Oct 15, 2022

Sounds interesting. Could you give a concrete example?

@Dsmasha
Copy link
Author

Dsmasha commented Oct 16, 2022

Sure. So for one image I use the prompt "1boy, solo, cowboy shot, clothes, {city|mountains|forest}" and checking the generated image afterwards shows the prompts as "1boy, solo, cowboy shot, tank top, shorts, sneakers, city" and the image was generated with only a city background as it should be. However, if I take "{city|mountains|forest}" and place it on the same lines as the clothes in clothes.txt, the image generates with the prompts "1boy, solo, cowboy shot, tank top, shorts, sneakers, {city|mountains|forest}" and contains a mix of all three backgrounds.

I tried to attach the generated images as examples, but it wouldn't let me.

adieyal added a commit that referenced this issue Oct 18, 2022
See this issue for a use case: #3
@adieyal
Copy link
Owner

adieyal commented Oct 18, 2022

Sure. So for one image I use the prompt "1boy, solo, cowboy shot, clothes, {city|mountains|forest}" and checking the generated image afterwards shows the prompts as "1boy, solo, cowboy shot, tank top, shorts, sneakers, city" and the image was generated with only a city background as it should be. However, if I take "{city|mountains|forest}" and place it on the same lines as the clothes in clothes.txt, the image generates with the prompts "1boy, solo, cowboy shot, tank top, shorts, sneakers, {city|mountains|forest}" and contains a mix of all three backgrounds.

I tried to attach the generated images as examples, but it wouldn't let me.

Just implemented it, could you give it a whirl and let me know if it works for you?

@Dsmasha
Copy link
Author

Dsmasha commented Oct 20, 2022

Just gave it a try and it works perfectly! Thanks so much.

@Dsmasha Dsmasha closed this as completed Oct 20, 2022
@adieyal
Copy link
Owner

adieyal commented Oct 20, 2022

That's great. Hope it's useful.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants