varunshenoy / opendream Goto Github PK
View Code? Open in Web Editor NEWAn extensible, easy-to-use, and portable diffusion web UI ๐จโ๐จ
License: MIT License
An extensible, easy-to-use, and portable diffusion web UI ๐จโ๐จ
License: MIT License
easier for users. we can just grep all of the files
add a more comprehensive README (outlining design decisions)
In case you create a mask layer that you don't like, for instance
Should be written this requires npm.
Might be missing other requirements? I'm still trying to get this running and hitting issues where it seems like no extensions are loaded.
for SAM/masking, Kayvon mentioned this
not that hard to do
Should be able to see prompt etc and experiment with slight changes.
Options should be to "create new layer with changes" or "update layer"
Should not be sending empty strings
Just use openCV or something to perform ops such as (increase saturation, gray out, zoom, sharpen, etc.). Show a good mixture of stable diffusion and basic image processing operations.
Hello. I just came across this project and it seems interesting.
Perhaps a simple "hello world" style workflow demo would be helpful. You know showing something like a screencast of running the app, installing a new model, a new extension, and showing how to update if possible. Also it would be helpful to also show what the app is capable of. Reading around the repo, I'm not sure how the app works, or how to install models, if that is at all possible to begin with.
I realize most projects have limited manpower to do multiple things, but demos and plenty of docs are crucial to encourage people to try things out.
Anyway,
Best of luck
I am beginner in this AI world, trying to get this UI running in ec2 instance but having some difficulties getting it to work. I have managed to launch the UI using sh run_opendream.sh
and get the UI loading in port 3000. However, I cannot get anything else working other than loading the UI. Here is the error shot from the inspect element, it seems to point something at port 8000. How do I get this working in ec2?
Use case: network produces several images. Code currently places all images in one layer. We don't want this.
How can I load local models/extensions easily like webuiautomatic1111, the workflow is great, but when I start and run, I can use very little, is there any tutorial, thank you๏ผ
Similar to .ipynb, we want extensions to be baked into the workflow that is exported.
Attempted to add an instruct_pix2pix
layer using the default install.
Nothing in the logfile other than the above error.
INFO: 127.0.0.1:61305 - "GET /schema/instruct_pix2pix HTTP/1.1" 200 OK
INFO: 127.0.0.1:61305 - "POST /operation/instruct_pix2pix/ HTTP/1.1" 307 Temporary Redirect
'str' object has no attribute 'get_image'
INFO: 127.0.0.1:61305 - "POST /operation/instruct_pix2pix HTTP/1.1" 500 Internal Server Error
Mac M2, latest main
commit.
workflow.json
: https://gist.github.com/mmastrac/15d1f94100a8b312fb96d0dae6ca063e
how do we create a mask? follow user mouse? etc
more generally fix these
and SAM allow for semantic masks
should allow for options to download in various formats/resolutions or maybe simple stub/default for now
Easy way to get setup
operators should return multiple layers by default (not just one)
Hello,
Is it possible to add SD XL support?
I'd love a colab notebook for this as well.
Thanks!
generating mask for (758, 876, 4)
set_torch_image input must be BCHW with long side 1024.
with body.png. test.png works fine.
Once a task is given, the modal should disappear and push the job onto some queue on the backend. This way you can launch a bunch of diffuse tasks at once.
There's a lot of places in the code where we implicitly/explicitly support only 512 x 512 images. If we want to add tools like "generative fill", we need to drop this dependency.
Create a simple Python extension file that can handle Stable Diffusion with LoRAs. It might make sense to create a separate command (not dream
).
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.