Replies: 2 comments 2 replies
-
Hi! There are a lot of things that are already controllable. For your request: I also want to add more concepts, themes, subjects to the csv files, or create more templates. Check here how to add things to csv's without breaking things on updates: Check here how to control certain parts of the generation process (such as randomness and subjects etc): As for a standalone application. I am thinking about this, but I currently don't have a good development environment (some life things going on!). I might be able to do more in a couple of weeks. I do have another idea to make this available somewhere else on the web, but I need to find the time for it. I thought about using LLM's, but I don't think it is worth it. The whole idea is to discover new and unexpected results. There are already prompt generators that use a model in the background. Check out Prompt Magic in dynamic prompts extension! https://github.com/adieyal/sd-dynamic-prompts#magic-prompt |
Beta Was this translation helpful? Give feedback.
-
Hey there, could you please tell me what I can change in the code so I can run the project as standalone ? Without the need to open Automatic111. Please. |
Beta Was this translation helpful? Give feedback.
-
Hi, I want to ask. Could you possibly make this tool as a standalone application for desktop ? I you cannot, would it be possible to tell me how to do it ? I really love this tool, but I don't want to open automatic1111 all the time. I want to have my prompts to a file. I also want to add more concepts, themes, subjects to the csv files, or create more templates.
You did great work and congratulations. If you could please answer this, because I would really like to have it as a standalone application. It does not need to be an .exe it could work with gradio. I just want the tool to work alone so I can add more things to it.
Thanks in advance.
(I managed to get the build_dynamic_prompt() function to work on Spyder. I also add more stuff to the csv files. But I cannot control the variables, for example to turn on or off specific art styles or subjects. I also tried to understand how the templates work but I didn't figure it out. My programming skills are limited even though I want to tweak this tool. Maybe you could help me.)
Also another thought : I thought of putting many prompts generated by OneButtonPrompt into ChatGPT and asking it to correct them to be more coherent and aesthetically pleasant. I didn't manage to do it as I want. Anyway, I suppose that we could feed many prompts into an LLM and then finetune it to produce better prompts. It is just an idea.
Beta Was this translation helpful? Give feedback.
All reactions