Comments (8)
Second that, it's kinda limiting after all for no apparent reason....
In the meantime, you can just change it here ;)
Please give back some flexibility, it's nice to be able to just step back and wonder what's coming out of that batch :D
https://github.com/hlky/stable-diffusion/blob/e2897e6b17579b499578d2e88f6337025adaae08/frontend/frontend.py#L34
from stable-diffusion.
Thanks! That's super helpful.
from stable-diffusion.
I agree @altryne lowered it to 10 and it's a bit low. It was 250 before I believe, however some users reported the user interface freezing after a batch of 200 completing, I think they were just being impatient, we will likely increase the batch size slider upper bound to a more sensible limit that won't freeze the gallery, and add a second option with a text box to enter a number but that will then disable gallery output, could probably add a progress bar or something instead though
from stable-diffusion.
It's hard for us to access the usage of features as we have no analytics (@hlky I'll add some basic ones)
@JamesIV4 @Aishou you guys can override the limits now with dynamic params in the prompt. This was announced here :
Sygil-Dev/sygil-webui#422
so if you type numer:100 in your prompt it will allow you to go out of the boundaries of the defaults easily without changing code
from stable-diffusion.
The issue with a LOT by default was twofold:
- was really hard on smaller screens to nudge the slider from 4 to 5 for example, and many folks don't do HUGE batches
- The batch size and batch count were confusing and folks tried and got a CUDA error cause.. it's heavy, and continued opening tickets
- Gradio UI doesn't give an option to add arbitrary values to sliders (it's on open issue of theirs)
from stable-diffusion.
We don't need analytics they wont tell us anything issue reports can't. Don't add any analytics.
from stable-diffusion.
@JamesIV4 batch count was increased to 50
from stable-diffusion.
50 is great, and nice to know about the override, very useful 👍
from stable-diffusion.
Related Issues (20)
- No Image Showing After Generation
- [BUG] When SD is relaunched it switches port numbers, it should maintain the same port to preserve user parameters
- ReadMe update - using GFPGAN reduces available memory
- [BUG] When attempting to use the img2img functionality, the server outputs an error instead of generating a new image. HOT 1
- [Bug]ZeroDivisionError: division by zero. No i2i result since yesterday update HOT 4
- Resize mode in img2img : tensor size error HOT 7
- Seed number is not precise for single txt2img image generated in batch HOT 3
- Not able to regenerate exact result from yaml file / some info is missing HOT 1
- Repeated relaunches upon first run after Linux installation
- CPU Support HOT 2
- New error on start after latest update. HOT 1
- k_euler very slow suddenly HOT 1
- Newest release broke webui IMG2IMG HOT 1
- Aspect ratio box seems to be broken HOT 2
- Poor text responsiveness in prompt input box after recent update HOT 29
- unsupported operand type(s) for +=: 'dict' and 'str' HOT 7
- Script for AMD? HOT 7
- Expose API HOT 2
- Personalization (textual inversion) seems broken with WebUi
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from stable-diffusion.