Spaces:

kadirnar

/

Video-Diffusion-WebUI

like 261

Runtime error

hi @chrka ,
You probably want to duplicate the Space and run on your own GPU

Change this parameters to increase the num of frames and steps
https://huggingface.co./spaces/ArtGAN/stable-diffusion-2-infinite-zoom-out/blob/debc5f37412b1317fa8bb3748528c1d9bc307749/inpaint_zoom/app/zoom_in_app.py#L84
https://huggingface.co./spaces/ArtGAN/stable-diffusion-2-infinite-zoom-out/blob/debc5f37412b1317fa8bb3748528c1d9bc307749/inpaint_zoom/app/zoom_out_app.py#L120-L126

thank you so much ! ... is there a step by step how to duplicate and run on my own gpu?
i have coding experience, also all that here is very new for me.

if you have your own GPU/CUDA env, you can just clone this repo

git clone https://huggingface.co./spaces/ArtGAN/stable-diffusion-2-infinite-zoom-out
cd stable-diffusion-2-infinite-zoom-out
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python app.py

Very cool app, I love the results!
So if I duplicate space in order to update the parameters (frames/steps), would the minimum be a T4 Small that this would run on? Thanks!

thank you very much.
i still have 3.11 installed... (ur example needs <3.11)
i installed 3.10, but it still shows --version = 3.11

how would i remove 3.11 and make 3.10 actual

next i started my 3.10 manually
with
git clone https://huggingface.co./spaces/ArtGAN/stable-diffusion-2-infinite-zoom-out
cd stable-diffusion-2-infinite-zoom-out
python3.10-intel64 -m venv venv << here i call it manually
source venv/bin/activate
pip install -r requirements.txt
python app.py

now i did not got the first version error... but this one

Collecting torch==1.13.1
Using cached torch-1.13.1-cp310-none-macosx_10_9_x86_64.whl (135.3 MB)
Collecting xformers
Using cached xformers-0.0.16.tar.gz (7.3 MB)
Preparing metadata (setup.py) ... error
error: subprocess-exited-with-error

× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> [6 lines of output]
Traceback (most recent call last):
File "", line 2, in
File "", line 34, in
File "/private/var/folders/6l/ywbg8w_x27lbk5dgj_t73k8w0000gn/T/pip-install-laizdd5b/xformers_19e155bf09e6499bbda451cf21423278/setup.py", line 23, in
import torch
ModuleNotFoundError: No module named 'torch'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

[notice] A new release of pip available: 22.3.1 -> 23.0.1
[notice] To update, run: pip install --upgrade pip
Traceback (most recent call last):
File "/Users/christianmueller/stable-diffusion-2-infinite-zoom-out/stable-diffusion-2-infinite-zoom-out/app.py", line 1, in
from inpaint_zoom.app.zoom_out_app import stable_diffusion_zoom_out_app
File "/Users/christianmueller/stable-diffusion-2-infinite-zoom-out/stable-diffusion-2-infinite-zoom-out/inpaint_zoom/app/zoom_out_app.py", line 1, in
from diffusers import DiffusionPipeline, DPMSolverMultistepScheduler
ModuleNotFoundError: No module named 'diffusers'
(venv) christianmueller@Christians-MacBook-Pro-2 stable-diffusion-2-infinite-zoom-out %

sorry for headache... :)

ohh @chrka it looks like you're on OSX, I don't think there's support for Mac yet

really? :(
anyhow., can i purchase server power here on huggingface and just clone it here? or what would be the simpliest option for me.
i have a dedicated server in the net for another project, also as said i am not so much in terminal and that stuff. as easier, as better.
i just like to create my own training, and using awesome apps like u offer and adjust.

thanks buddy for help

@chrka ,

mps_device = torch.device("mps")

For mac you have to add this code.
Doc: https://pytorch.org/docs/master/notes/mps.html

omg you are awesome, thank you guys.
also i am still lost... where to add all this sourcecode?

thats what i got from @radames
git clone https://huggingface.co./spaces/ArtGAN/stable-diffusion-2-infinite-zoom-out
cd stable-diffusion-2-infinite-zoom-out
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python app.py

kindly tell me how to include your steps into that and where to add.

thank you

btw if anyone is interested for some fee to help me in my endeavors, kindly let me know.
my email: [email protected]

i would like for example to install a chatgpt chat on a webpage on my dedicated server.
that i can either pre-trained further, or at least add some extra infos about the user in the prompt.
that extra infos come from a mobile app i have via post .
it should be in a mobile app size format and maybe a little menu where the users can add their
own chatgpt api key and choose from a few predefined prompts.
i saw some source here: https://github.com/orhanerday/ChatGPT
also, as much i coded in the past, i feel its time to give that job to some youngsters :)

and there are other projects , with vr, ai, nft.

thanks for listening, that was a lot of infos :)

christian

@chrka ,

Diffusers Mac: https://huggingface.co./docs/diffusers/optimization/mps
I will write a webui application for Text2Text models.

I will add Mac support in 1-2 week.

awesome.
in the meantime, i am going forward to setup my visual studio code environment with my namecheap and ssh,
installed already python 3.9. still have a challenge to connect with ssh from visual code to my namecheap (while it works in the terminal)... any idea? :)

whatever works first, my mac or that remote server ...

thanks for your support

btw. i have no m1/m2.
its a macbook pro from 2017

Why it's not working in huggingface online, model is loading endless. Is it bug or this model just for download?

Can you check?

Now it's working. It's not worked, because someone click on update button near the models selector (maybe I XD). If you click update the program, it will check for all diffusion models, so it can take a really big time.

Man, make please outputs as temporary files. It will help to solve the errors with outputs and not to load model storage.

I don't write the codes. I am using github repos. I will reduce the number of models.

That's good idea, I thought about it. Thank you!