imrayya/stable-diffusion-webui-prompt_generator

An extension to AUTOMATIC1111 WebUI for stable diffusion which adds a prompt generator

stable-diffusion-webui-prompt_generator's Issues

Extremely slow model download

One of the models, the 304 MB large one, seems to download from a very, very slow server at 30 KBps on a 25 MBps connection. I couldn't find the part that downloads it so I don't know where it's fetched from but it would make sense to rehost it on Huggingface if the license allows for it.

UI

Hello, I am using the integration package webui provided by others. After installing the plug-in, the UI is not displayed correctly. How to solve it?

Export to file

Would be nice to include a export to file, or option to export as file.

Blacklist not working

This extension is so cool! Thank you for making it. I noticed the blacklist does not seem to be working. For example, I have "unreal" and "emma watson" in there, but I am still getting prompt results containing those words. I have tried restarting the app after saving the list.

[Bug] Casting top_k and penalty variables

coming from the automatic1111 implementation of this repo. I noticed the repetition penalty and top_k variables aren't casted to floats and integers, respectively, which will cause error when the average user dragging the sliders; fyi

When I press the "send to txt2img" button, an error appears

generation, everything works, but there is an error when sending.

Artist names

Is there a way to limit the artist names being generated?
This is such a helpful extension as I'm a big fan of anything introducing randomness and saving on typing but I see the same artists all over the internet and don't want to join in with this trend.
I know it's not your model but is there a way to opt in or out of using artists in the prompts?

I can't install in the webui or even load after downloaded manually

As the title said, I can't install in the Extension tab. The error says:

Error executing callback ui_tabs_callback for E:\AI_RND\stable-diffusion-webui-master\extensions\stable-diffusion-webui-Prompt_Generator-master\scripts\prompt_generator.py
Traceback (most recent call last):
File "E:\AI_RND\stable-diffusion-webui-master\modules\script_callbacks.py", line 103, in ui_tabs_callback
res += c.callback() or []
File "E:\AI_RND\stable-diffusion-webui-master\extensions\stable-diffusion-webui-Prompt_Generator-master\scripts\prompt_generator.py", line 160, in on_ui_tabs
populate_models()
File "E:\AI_RND\stable-diffusion-webui-master\extensions\stable-diffusion-webui-Prompt_Generator-master\scripts\prompt_generator.py", line 34, in populate_models
with open(path, 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: './extensions/stable-diffusion-webui-Prompt_Generator/models.json'

Even when I manually downloaded the git and extracted to extension folder, the A1111 still got error.

Broken Tool Tip

It took some time trying to figure out why my tooltips were broken in AUTOMATIC1111

stable-diffusion-webui

After exhausting all community posts to try and resolve the problem I unloaded all of my extensions and fixed my tool tips. I started loading extensions 1 by 1.

I discovered that this extension breaks the tooltips.

I don't like making bug reports so I hope this finds you in good spirits.

results does not appear with latest automatic 1111

Hi @imrayya your extension worked pefectly, but
since upgrading automatic1111, results and buttons are not shown, even with default theme, and no other extension, even after prompt_generator reinstall

python: 3.10.9  •  torch: 2.0.0+cu118  •  xformers: 0.0.17  •  gradio: 3.28.1  •  commit: [5ab7f213](https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/5ab7f213bec2f816f9c5644becb32eb72c8ffb89)  •  checkpoint: [92970aa785](https://google.com/search?q=92970aa785eb76e427847109a8f4ec6abfab36ef941f78d295d323d79f6130c9)

Do you have an idea why ? thxs

with no other extension,

Latest A1111 broken it?

Latest A1111 pull seems to cause issues for me with extension:

Exception encountered while attempting to generate prompt: local variable 'tokenizer' referenced before assignment
Traceback (most recent call last):
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\routes.py", line 321, in run_predict
output = await app.blocks.process_api(
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1016, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 945, in postprocess_data
if predictions[i] is components._Keywords.FINISHED_ITERATING:
IndexError: tuple index out of range
Exception encountered while attempting to install tokenizer
Generate new prompt from: "Basketball, nba, landscape, rim, 35mm, 8k"
Exception encountered while attempting to generate prompt: local variable 'tokenizer' referenced before assignment
Traceback (most recent call last):
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\routes.py", line 321, in run_predict
output = await app.blocks.process_api(
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 1016, in process_api
data = self.postprocess_data(fn_index, result["prediction"], state)
File "E:__workspace\stable-diffusion-webui\venv\lib\site-packages\gradio\blocks.py", line 945, in postprocess_data
if predictions[i] is components._Keywords.FINISHED_ITERATING:
IndexError: tuple index out of range

Send Batch Results to Prompt From File

Describe the solution you'd like
Provide the feature to send a batch output to the "Prompt From File" script available in Text2Img and Img2Image.

Additional context
Reference to #24

Hard written path

Line 41 of the script prompt_generator.py, the directory name "stable-diffusion-webui-Prompt_Generator" should not be hard written.
__file__ might be used, to allow changing the ext directory name.

Request: Send prompts to text2img/img2img and generate negative prompts

can we have a way to automatically send a prompt to to text2img/img2img etc without copying and pasting?
and can it also generate negative prompts?

Does this extension download the models or does it send prompt requests to some endpoint?

I'd appreciate if you could make this clear somewhere on the repo... does your extension download the models locally or does it make inference requests to some free-access endpoint ?

I could read the code myself and be able to tell, but I think those who can't should also be informed about this.

Feature Request: "Send to X" buttons for prompts.

#19 made me think of this...

Can we please have buttons? :-P

Include the Parameters Explanation paragraph in the UI

I think it would be great if you could add the Parameters Explanation paragraph somewhere in the extension UI... it would be very convenient to be able to read it right there.