KaliYuga's Library of AI Google Colab Notebooks [Updated Mar 11, 2022]

in The Latent Space3 years ago (edited)

Hi guys! As some of you know, I've been involved in creating AI art for almost two years. When I'm not building my own models, I use a lot of different Google Colab notebooks in my art. Even though there are a lot of really excellent notebooks available, tracking them down can be a challenge. I haven't seen a really comprehensive list of the ones used by the nascent AI art community, so I figured I'd put one together and give more people the freedom to create. This post will continue to grow as I find new notebooks, and as new methods are created. If you find any I've missed or have created any, let me know and I'll add them!

Table of Contents

  1. The Big Sleep
  2. VQGAN + Clip
    2a. VQGAN+CLIP with Video Features
  3. Clip-Guided Diffusion
  4. Styledream Notebook
  5. Pixray PixelDraw
  6. Real-ESRGAN Practical Image Restoration
  7. ruDALLE [Added Nov 2, 2021; Updated Nov 3, 2021]
  8. Looking Glass AI [Added Dec. 28 2021]
  9. JAX Clip-Guided Diffusion *[Added Mar. 11, 2022]
  10. CLIP Guided Deep Image Prior *[Added Mar. 11, 2022]
  11. Disco Diffusion [Added Mar. 11, 2022]
  12. Disco Diffusion Turbo [Added Mar. 11, 2022]

1. The Big Sleep

image.png
A Beautiful House by BigSleep

Description: The original Clip-guided AI Colab tool; uses Clip to guide BigGAN. Fairly outdated, but absolutely foundational to the entire art field.
Author: Ryan Murdock (advadnoun on Twitter)
Notebook Link


2. VQGAN + Clip

image.png
Gordian Gallery by KaliYuga, 2021

Description: By far the most widely-used of the Clip-guided methods [Edit April 2022: very few people still seem to use this tool, but it's a foundational and very important one]; uses Clip to guide VQGAN.
Author: Katherine Crowson (RiversHaveWings on twitter)
Notebook Link (The notebook is in Spanish, but is easily translated to English using Chrome).


2a. VQGAN+CLIP with Video Features

Description: What it says on the tin! VQGAN + Clip with added video features and also some really good demystifications of parameters.
Author: @jotakrevs
Notebook Link


3. Clip-Guided Diffusion

image.png
Psychic Sermon in the Temple of the Mind by KaliYuga, 2021

Description: Produces absolutely unbelievable results, but very slow and resource-heavy. Uses Clip to guide a Diffusion model.
Author: Katherine Crowson (RiversHaveWings on twitter)
Notebook Link


4. Styledream Notebook

image.png
Portraits of DC heroes in the style of Alex Ross *by .hack//HAMBRE

Description: Portraits only! Clip-guided StyleGAN model.
Author: Ariel Ekgren (ArYoMo on twitter)
Notebook Link


5. Pixray PixelDraw

image.png
A synthwave great white shark by Viktor Alm

Description: A fork of NerdyRodent's fork of RiversHaveWings'VQGAN + Clip Notebook; uses VQGAN + Clip to produce pretty impressive pixel art.
Author:Dribnet
Notebook Link
Extra! You can also start with an initial image in a different version of Pixray; if you use an image originally created by pixray, the results can be pretty trippy! For example:

image.png
Via the Colab Notebook's demo params

Other bits and pieces of this project can be found here.


6. Real-ESRGAN Practical Image Restoration

image.png
Examples via the Colab Notebook

Description: A demo of Real-ESRGAN,which repairs many types of image degradation and also upscales the image 4x.
Authors: Xintao Wang, Liangbin Xie, Chao Dong, Ying Shan
Notebook Link


7. ruDALLE

image.png

Description: A Fully-Trained, Accessible, Multi-Billion-Parameter Answer to OpenAI's DALL-E Model (in Russian)
Authors: Sber AI and Sber Devices
Notebook Link
Bonus: A faster Colab notebook for free users


8. Looking Glass AI

image.png
"Untitled" by KaliYuga, 2021

Description: An Accessible One-Shot Finetuner For RuDalle
Authors: AI_Curio and Others
Notebook Link


9. JAX Clip-Guided Diffusion


By KaliYuga, February 2022

Description: [Via Gituhub] "The models are denoising diffusion probabilistic models (https://arxiv.org/abs/2006.11239), which are trained to reverse a gradual noising process, allowing the models to generate samples from the learned data distributions starting from random noise."
Authors: Katherine Crowson and John David Pressman
Notebook Link


10. CLIP Guided Deep Image Prior

image.png
By KaliYuga, February 2022

Description: Creates soft, print-like images quite unlike what you generally get out of other CLIP-guided diffusion models
Authors: Daniel Russell, based on work by Katherine Crowson and Ryan Murdock
Notebook Link


11. Disco Diffusion

image.png
By KaliYuga, March 2022

Description: My daily drive; Very high quality image results very quickly; comes fully loaded with 3D animation capabilities and the latest Clip models.
Authors: Somnai and Adam Letts
Notebook Link


12. Disco Diffusion Turbo

~~~ embed:1502057904693723137?s=20&t=MLWq2-K9-Rc7Bu5IN9wZWg twitter metadata:S2FsaVl1Z2FfYWl8fGh0dHBzOi8vdHdpdHRlci5jb20vS2FsaVl1Z2FfYWkvc3RhdHVzLzE1MDIwNTc5MDQ2OTM3MjMxMzd8 ~~~

Description: A fork of Disco which speeds up and cleans up 3D animation significantly. I also use this all the time!
Author: Chris Allen
Notebook Link
Bonus by Chris Allen Notebook: Wiggle Standalone Keyframe Generator; lets you generate keyframe instructions that give you fine control over Turbo's motion.

Sort:  

By the way, I think @juecoree should see this, and join our boat (:

Thanks for the heads up! I'll definitely check out and try this list of tools.

This post was perfectly timed. I was just about to experiment and play around with some AI art and came to look up a post for links and found this! Thanks for the in depth post and all the invaluable resources! It's much appreciated! =) Awesome art outcomes too! Keep up the great work. =)

Hey! Glad you found it helpful--make sure to post what you come up with :)

I have been slacking on posting. I have a folder for each day's worth of work for about 20 or more days. So, i have a lot of generated content... i just haven't posted or shared yet. I need to get busy and just put some out there. I hate the part of me that wants to be a perfectionist... hahaha it doesn't suit me. Thanks for the encouragement and interest! Cheers. =)

Just what I was looking for and was in fact planning to include whatever notebooks I knew of in my next post. Will bookmark this post, thank you for the comprehensive list :)

Oh this is awesome! You've here some interesting resources.

By the way, thanks for the attribution in the "Video Features" notebook, but I just fixed some outdated URLs and a bad typo. These people are who deserve the attribution, not me (:

Generate images from text phrases with VQGAN and CLIP (z + quantize method), with animation and keyframes
Notebook by Katherine Crowson (https://github.com/crowsonkb, https://twitter.com/RiversHaveWings). The original BigGAN + CLIP method was made by https://twitter.com/advadnoun. Translated into Spanish and added explanations, and modifications by Eleiber#8347, and the friendly interface was made thanks to Abulafia#3734. Translated back into English, and zoom, pan, rotation, and keyframes features by Chigozie Nri (https://github.com/chigozienri, https://twitter.com/chigozienri) If you encounter problems using it, you are welcome to ask me to fix it at https://twitter.com/chigozienri

Slight modifications by : Justin John

Oh, gotcha—I’ll update that :)

This is excellent. Using VQGAN + Clip

Thanks for creating this list; in terms of 11 and 12 in your inventory (Disco Diffusion), these have now been merged, and additional forks added to the main notebook to form 5.4 - https://colab.research.google.com/github/alembics/disco-diffusion/blob/main/Disco_Diffusion.ipynb