81–90 of 148 posts
I’m waiting for someone to wrap this up into a desktop app that I can install and run on my Mac.
There are a few bugs to iron out before it's ready for prime time. For now, create the folder `~/Desktop/charl-e/samples/` manually before you run it.
a sigh of relief as I thought that this would generate instead of pictures, React UI code based on plain text description. First they came for illustrators, then they came for UI designers.
Earlier quoted context omitted.
Colab: https://colab.research.google.com/github/WASasquatch/StableD... To run it elsewhere in the cloud, grab a GPU (spot) instance and SSH in.
You mean a VM on a machine with a GPU? Or does it have to be a bare metal machine? What is a good provider of suitable VMs/machines? And what do you do after you SSHed in? The installation instructions seem to be for windows users (click here, then click there ...) is there a linux script that does the installation automatically?
Parent comment alludes to docker-compose up.
Earlier quoted context omitted.
> As it’s deterministic (given the exact same request parameters, random seed included, you get the exact same image) it’s a form of compression (or at least encoding decoding) too: I could send you the parameters for 1 million images that you would be able to recreate on your side, just as a relatively small text file. For any input image? Or do you mean an image generated by the model?
You could input an image and get it to recreate it as best as possible and then output a seed. That would be interesting!
After running for a while, the adversarial network outputs a seed, and you now have a few characters representing a reasonable approximation of your image.
Earlier quoted context omitted.
You mean a VM on a machine with a GPU? Or does it have to be a bare metal machine? What is a good provider of suitable VMs/machines? And what do you do after you SSHed in? The installation instructions seem to be for windows users (click here, then click there ...) is there a linux script that does the installation automatically?
Just a VM with a GPU, doesn't need to be bare metal. AWS/GCP/Azure has em, but for GPU cloud instances, boutique vendors like CoreWeave, runpod, lambdalabs.com, vast.ai, paperspace may be more competitive. Parent comment alludes to docker-compose up.
Earlier quoted context omitted.
I've been looking into this for the last 2 days. Unless you're running an M1 Mac or newer, you're SOL. Stable Diffusion is built on PyTorch. PyTorch mainly has been designed to work with Nvidia cards. However PyTorch added support for something called RocM like a year ago that adds compatibility with newer AMD cards. Unfortunately RocM doesn't support slightly older AMD cards in conjunction with intel processors. So…
There is work on a CoreML version which may play nicer with older Macs w/sufficiently beefy dGPUs. https://github.com/huggingface/diffusers/issues/443
Earlier quoted context omitted.
Just a VM with a GPU, doesn't need to be bare metal. AWS/GCP/Azure has em, but for GPU cloud instances, boutique vendors like CoreWeave, runpod, lambdalabs.com, vast.ai, paperspace may be more competitive. Parent comment alludes to docker-compose up.
You can run docker inside a VM? And it will be able to use the GPU and tunnel port 80 through the docker container, through the VM and to the web?
Earlier quoted context omitted.
My understanding is that images will not be bit identical due to GPU physics and decimal precision. Images from the same seed may be for all practical intents and purposes indistinguishable - but there are some flipped bits involved.
That's not my understanding. The same seed value to the device's random number generator should results in the exact same outputs - there's a bug being chased down in the MPS (MacOS) backend where the fixed random seed doesn't output the same image on different computers.
No idea how true that is, but on my windows machine, same params/seed is definitely deterministic.
[0] a help string in the SD source code recommends the ddim_eta parameter (which isn't exposed in most web UI or GUI's, including the OP github) stay at the default 0.8 for deterministic sampling. I have no idea if this means changing the value from 0.8 produces non-deterministic results with the same hardware/os/params/seed. Or if they just mean changing this from 0.8 will make your SD not match the online model but still be deterministic itself. But in my testing, changing this value gives no useful changes to the image generation, so I keep it at 0.8
Earlier quoted context omitted.
You could input an image and get it to recreate it as best as possible and then output a seed. That would be interesting!
This is a fascinating idea. Have StableDiffusion generate an image from the image you'd like to "compress" + a random seed. Feed that output to an adversarial network that compares source image to output and scores it. Try again with new seed. After running for a while, the adversarial network outputs a seed, and you now have a few characters representing a reasonable approximation of your image.