theaiquestion

There's a ton of different methods of multicolor printing each with different pros and cons - simplified

* nozzle that mixes from multiple inputs * multiple hotends on the carriage * swappable hotends * single hotend - retracts and switchs to a different filament, few methods here, this is what the AMS and pallete are **

You can do it with multiple heads, the problem is ooze. You'll need to either be able to retract the non-used printhead so that it doesnt ooze onto your print, or be able to swap it mid print. Either way here you'll still need a method to prime it. Some materials may break down in the heat of the hotend too while they sit there unused just baking.

There's some methods that make the blocks smaller, like using the infill as the priming section.

Because most company's/people are (AFAIK) under the impression that training on public data will fall under "Fair Use" because it's substantially transformative, in the case that it isn't then you've already agreed to it on github.

It's a fallback clause, "fair use" is irrelevant if you've already given github permission to use it. By adding that clause you can no longer argue that it's not fair use to use the code you put on github after agreeing to their terms.

Funny enough the first one I've found that works consistently with this one(however unusably destructive) is the "whirl" filter in GIMP.

Previous adversarial methods I've beat with simple noise/blur. This one is fairly resistant to basic filters.

Just for context LAOIN's largest dataset is 5.8 Billion images

And you'd be spending that compute time for what will be a tiny, tiny,portion of that

Did you switch to a mac or to a windows machine?

Yeah I was in school and wanted to try making games - but that was the exact year that "flash is dead - HTML5 is going to replace it"

And.... Yeah HTML gave us the canvas? No editor, no animations, no input.

The real innovations were with ES5/2015. It utterly destroyed the "mess around and make some basic games".

It's utterly laughable that "html5" was somehow the "alternative". We went from an editor that anyone can quickly create animations, games, etc to nothing - for over 10 years! And yeah to my knowledge we're _still_ not back to what we used to have.

Has any of the OSINT/INFOSEC crowd moved to mastodon or somewhere other then twitter?

I think the method of merging the pipelines via img2img should use controlnet. Possibly needing to be finetuned specifically for this, although existing controlnet models might work fine for this.

This is exactly what you'd want to use controlnet for - mapping semantic information onto the perceived structure.

Delaying public usage by decades?

I don't mean to be rude, and to be clear I wish they did release the weights, but what did they lose here?

You don't approve - so what? Releasing the weights doesn't make them money.

I'm skeptical LLaMa is even useful for facebook commercially at least, they don't make money on it, and I doubt anyone developed brand loyalty, more then likely everyone will use whatever the next, best open model is regardless of who makes it.

Llama and SD still don't come close to midjourney/chatgpt/claude when you look at ease of use and infrastructure cost. These "99% the performance of chatgpt" are laughable if you use them (which I have extensively).

> We are done inhaling vapor.

Okay? What were you about to pay for to begin with here?

EDIT: Just to add, it's not like we got nothing from this, this can likely still be something to try with SD.

> How. Many. Times. Have we heard it already?

AFAIK what they've been saying so far has at best been "we're able to continue making them at our current pace". They've also been saying they were focusing on commercial partners over consumers (which IMO is better because commercial customers need some level of assurance of future stock to actually use pi's in their designs).

Neither of those things particularly screams "Regular customers will soon be getting pi's in their hands"

December 2022 They said they expected better stock this year, which appears to be accurate.

I'm more curious to see if they can manage to make a PI 4 successor given how many years the pi 4 has been around already.

I feel like people don't understand what this is, and what google has historically done.

This is research for some math to improve the ability to do a style transfer, and they show it on their own text-to-image generator muse, which they have also published the structure of.

This is what they have typically done, and this is what they didn't do with Bard.

No, they did not release waits for it, you cannot run this on your own computer. But they typically didn't release weights for things like Lambda or Imagen either AFAIK.

This is not a product. This is not a tool for you to use. This is for researchers.

The point of this paper is not to let you run it on your computer. It's to allow other researchers to implement and build on the methods described in the paper.