bryced

It is stable diffusion but yes my fork does not have the memory optimizations needed to run it on only 4gb

> Another interesting (albeit much harder) thing to try would be to take the image output from SD, run it through MiDaS to get a new depth map with additional details. Diff the depth maps in 3d space and update the geometry to match before projecting the image. Combine it with the first suggestion and you would have a process for progressively refining a detailed, fully textured model.

I made a go at this exact approach last night. I'm new to working with 3d data but at least got a mesh rendered from the depth map. Spent most of my time fighting with tooling. Let me know if you want to help out.

Stable Diffusion 2.1 fixes the bug in 2.0 that caused people to be rendered poorly.

Example of a fun way to make desktop wallpaper (make sure you set the desktop to tile the image):

>> pip install imaginairy

>> aimg

imaginAIry> imagine --model SD-2.1 --tile "fruit salad" "blueberries" "colorful abstract art" "christmas desktop background" -r 4

Possibly. Haven't tried. In principle should be possible.

I don't think it works with anaconda on any OS.

You're probably noticing the newest sampler, which also works with 1.5.

Yeah I tried to get it running but it kept crashing with "out-of-ram" errors.

Good idea to version the notebook.

- text-based masks use a clipseg model. - the boolean mask logic is unique to this library - the face fixing is done by CodeFormer

All the same issues as migrating 1.5 to M1s. It went fast because I upgraded my existing codebase that had those fixes already instead of building of the new compvis one.

It might work on windows but I haven't tested it there.

Try out the pre-release like this:

`pip install imaginairy==6.0.0a0 --upgrade`

New 512x512 model supported with all samplers and inpainting

New 768x768 model supported with the DDIM sampler only

Not yet supported is the upscaling and depth maps.

To be honest I'm not sure the new model produces better images but maybe they will release some improved models in the future now that they have the pipeline open.

> Will AI lead to UBI? (56:09)

> Maybe. It'll either lead to UBI and utopia or panopticon that we can never escape from because the models that were previously used to focus our attention and service ads will be used to control our brains instead. And they're really good at that.

>So, you know, no big deal, just two forks in the road. That's the way we kind of do.

One kind of terrifying comment from the interview.

The Q&A was packed full of information. I wanted to make this more accessible if you don't have time to listen to a long recording.

Created with Whisper and a few hours manually editing the text.

Super fascinating Q&A with Emad about the future of AI and upcoming model releases.

oh that's great - thank you!

For code that doesn't have good pypi packaging I vendor it using scripts in the makefile

Thanks for the feedback!

the "alpha" is already built into the tooling and specified independent of the mask. IE. the stable diffusion inpainting takes hints from what you leave and "decides" what to keep

This python library does the same thing (but didn't get traction when I posted it yesterday):

https://github.com/brycedrennan/imaginAIry#automated-replace...

https://news.ycombinator.com/item?id=32887385

And I got the idea from here:

https://github.com/ThereforeGames/txt2mask

Which is using the model here:

https://github.com/timojl/clipseg

Clipseg is doing the hard part!

Really cool new feature I just integrated into this python library: AI object replacement. You just specify something like "replace the fruit with gold coins" and Bam! it happens. Makes editing way way easier. Have fun!

For python developers with appropriate hardware, this should be the easiest way to get started generating images.

Doesn't seem very mysterious.

When taken in context it looks like an innocuous tweet but to a twitter censorship algorithm that isn't aware of the context it looks like a command for someone to die.

Seems most likely you replied to this tweet and forgot. Your tweet makes sense in this context.

"They arent bothering to train current conscripts. Some Russian soldiers say they are given a weapon and sent straight in. Learn on the job." 11:29 AM · May 2, 2022

https://twitter.com/Johnrigsby128/status/1521195277033648129

yeah here is another one. why didn't they take down the whole seller? https://www.amazon.com/Charger-Braided-Charging-Syncing-Comp... edit: put correct link

1600 reviews, almost a perfect rating. The reviews and questions are about other products like a bread machine. I reviewed the product as 1-star but my review doesn’t show up. I reported this listing to amazon several days ago but it remains.

Minecraft has an add on that does this called “The Lost Cities”

“There are highways, bridges, tunnels, a subway system, tons of dungeons with spawners and loot and so on. ”

https://minecraft.curseforge.com/projects/the-lost-cities https://youtu.be/OZYZ0kV7Uw0

Seems very similar to the riprova library. I've found it very useful for asyncio web scrapers.

http://riprova.readthedocs.io/en/latest/ https://github.com/h2non/riprova

Its like this site didn't exist until a few minutes ago when it was posted here and on reddit. (https://www.reddit.com/r/sanfrancisco/comments/3n2lxn/i_have...).

- there are almost no search results for this blog on google - there is no way to identify the author - no linkedin - no twitter - no facebook - there is only one other blog post and it isn't linked to anywhere on the net - its easy to backdate a post.

I just received spam email using my logmein-specific email address.

CircleUp - San Francisco, CA

Fulltime Front-End or Full-Stack Engineer

https://circleup.com/

CircleUp is a platform for investing in small private companies.

We need engineer #3 to help us with the rapid growth of our business! If you’re ready to build a quality product that helps solve a pressing real-world problem, and you like working in a fun, creative environment that welcomes new ideas, come join us. Love of food trucks & free consumer products samples is a plus :)

Skills & Experience

- Expert-level proficiency in HTML5, CSS3, Javascript, jQuery, and AJAX.

- Python/Django application stack experience, basic backend coding

- Familiarity with rich client-side Javascript frameworks (Backbone, Knockout, AngularJS, Ember)

- Familiarity with mobile web development best practices (i.e. responsive design, progressive enhancement, graceful  degradation

Responsibilities - You’ll work on the full stack and own the front-end (as Engineer #3) on our small, agile team of engineers and designers.

- You’ll help solve engineering challenges around the elegant display of complex information, the social graph, the mobile web, scale, security, performance, and more.

- You’ll rapidly iterate on a product roadmap driven by lean startup principles.

Apply at jobs@circleup.com