181–190 of 199 posts
Earlier quoted context omitted.
You're right. On top of that, python dependency / package management is a joke.
Normally it's fine, but Apple have decided to force everyone into installing a conda python alongside normal python, which makes everything way more complicated than it needs to be.
Since it's a Mac app, I have to wonder if it could stick the prompt, steps, and guidance into the notes field of Get Info? I find I'm generating a lot of relatively low guidance (I'd love a 6.5 option) images and iterating on the prompts with an eye to what it's suggesting to the algorithm. As such I have no way to closely track what prompt was active on any output as it changes so often.
I strongly suspect the real merit of this approach is not the crowd-pleasing, 'set very high guidance on some artistic trope so it's forced to fake something very impressive', but rather the ability to integrate a bunch of disparate guidances and occasionally hit on a striking image. It's like the harder you force it into a particular mold, the more derivative and stifled its output becomes, but if you let it free associate… I'll be experimenting. Seems like getting the occasional black image shows you're giving it the freest rein.
Looking forward to 'image to image' a lot. I assume the prompt still matters, as it's fundamental to the diffusion denoising? Image to image means iterating on visual 'seeds'.
I've seen talk of textual inversion training: it would interest me greatly to be able to generate objects and styles and train a personal version of SD in a sort of back-and-forth iteration. The link to language is really important here, but so is the ability to operate as an artist and generate drawings, aesthetics and so on, to train the model. I did 440 episodes of a hand-drawn webcomic once, which had recurring characters and an ink-wash grayscale style I gradually developed. That means I have my own dataset, which is my own property, and certainly didn't make it big enough to make it into Stable Diffusion like say Beeple did.
Interesting times for the cybernetic artist. Basically computer-assisted hallucinatory unconscious, plus computer-assisted rendering. You could feed all of Cerebus (Dave Sim and Gerhard) into a model like this, panel by panel, and you'd probably get a hell of a lot of Gerhard out because so much of the panel area is tone and texture from him…
Earlier quoted context omitted.
Can it be built from source? (Am thinking if yes, then the seed feature should be easy to add).
The seed problem is in upstream GPU libraries provided by Apple. The community is working on solutions, but they are not easy.
Earlier quoted context omitted.
Nice UI. Which GPU are you using? Seems to be really fast. I have 3090 and it's not that fast. 4 images with 20 timesteps takes 7.4 seconds.
Thanks :) I generate one image in about ~3 seconds with the DDIM sampler, 20 steps, on a RTX 2080Ti (~8it/s). The video on the Patreon page is sped up as it's not very interesting to sit and watch renders haha. Although, some of the users who started using my UI weren't using the fork my app connects to, and were surprised it was a bit faster than what they were using before, so maybe you can give it a try. The repos…
Earlier quoted context omitted.
> But the second point here is also wrong: the whole reason these models are interesting is because they can generate things they haven't seen before No, it's essentially generating mashups of its training data, which can be very interesting. So a model that hasn't been trained on a lot of porn will of course do a very bad job at generating porn.
Mashups is the wrong way to think about it. It's generalizing at a higher level than texture / image sampling and it can tween things in latent space to get to visual spaces that haven't been explored by human artists before. It requires a good steer and prompting is a clumsy tool for fine tuning - it's adequate for initialization but we lack words for every shade of meaning, and phrase weighting is pretty clumsy too…
The very fact that the model is interpolating between things in the latent space probably explains why its images haven't been explored by human artists before: because there is a disconnect between the latent space of the model and genuine "latent space" of human artistic endeavor, which is an interplay between the laws of physics and the aesthetic interests of humans. I think these models know very little about either of those things and thus generate some pretty interesting novelty.
The fact that I can do this on commodity hardware on a 4GB model. A model that understands text and visual images, just absolutely blows my mind.
I almost feel like in a new future, a 100GB model may be able to offline handle speech -> text, video -> live scene graph. A robot that could base level physical understanding of our world like a 4 year old does. (objects, their relationship to other objects and behaviors)
Earlier quoted context omitted.
The seed problem is in upstream GPU libraries provided by Apple. The community is working on solutions, but they are not easy.
Can you elaborate? Isn’t the seed a simple int (or long int)?
Earlier quoted context omitted.
Mashups is the wrong way to think about it. It's generalizing at a higher level than texture / image sampling and it can tween things in latent space to get to visual spaces that haven't been explored by human artists before. It requires a good steer and prompting is a clumsy tool for fine tuning - it's adequate for initialization but we lack words for every shade of meaning, and phrase weighting is pretty clumsy too…
"It's generalizing at a higher level than texture / image sampling and it can tween things in latent space to get to visual spaces that haven't been explored by human artists before." The very fact that the model is interpolating between things in the latent space probably explains why its images haven't been explored by human artists before: because there is a disconnect between the latent space of the model and gen…
Aesthetic choices like colour and shapes and composition combine with literal representations, facial emotions, symbolic meanings and so on. AI art so far feels quite shallow by this metric, usually only hitting a couple of notes. But sometimes it can play those couple of notes very sweetly.
Earlier quoted context omitted.
Oof. I guess I should at least 16GB on my next Mac (I've been very happy with 8GB on my M1 Air for just about everything else, though).
RAM won't expected the speed much, mostly impacts how large images you can render (8GB would be limited to 512x512 if not smaller). Memory bandwidth and available computer cores on the GPU matters more when it comes to generation speed.
According to others, it’s about a minute or less with 16GB.