1/ What makes this @ChemicalScience front cover so special is how it was made.

It is probably the first-ever journal cover designed using a VQGAN + CLIP and the unreal engine trick!
#VQGAN #CLIP #GenerativeArt

More information in the thread ⬇️ Image
2/ The AI model (VQGAN + CLIP) generated most of the image using “enzymatic chemical reactions. green chemistry. advanced unreal engine” as input.
It’s interesting that you can recognise the lab with the blackboards, the floor and the “reactions”.
3/ I’ve added an enzymatic reaction from our publication (pubs.rsc.org/en/content/art…) on top of the image.

Before (AI generated) | after (final cover) ImageImage
4/ Adding “unreal engine” or “advanced unreal engine” to the text input prompt is important as it substantially increases the quality of the output image.

This trick was recently made popular by @arankomatsuzaki:
5/ I had tested it on a few chemistry related inputs. One of the inputs now made it on the cover of @ChemicalScience!
6/ There is an awesome notebook written by @RiversHaveWings that helped me generate the images.
colab.research.google.com/drive/1L8oL-vL…


Thank you very much, @RiversHaveWings!
Check out her profile, you will find lots of awesome images!
7/ More details on how to operate the VQGAN+CLIP model can be found here (@images_ai):
docs.google.com/document/d/1Lu…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Philippe Schwaller

Philippe Schwaller Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @pschwllr

28 Sep 20
Taking chemical reaction prediction models one step further in a great collaboration with Giorgio (@Giorgio_P_), a brilliant organic chemist!

A thread ⬇️1/N
A major limitation of current deep learning reaction prediction models is stereochemistry. It is not taken into account by graph-neural networks and a weakness of text-based prediction models, like the Molecular Transformer (doi.org/10.1021/acscen…).
How can we improve? 2/N
In this work, we take carbohydrate reactions as an example. Compared to the reactions in patents (avg. 0.4 stereocentres in product), carbohydrate contain multiple stereocentres (avg. >6 in our test set), which make reactivity predictions challenging even for human experts. 3/N
Read 14 tweets
5 Mar 20
Awesome! All the video recordings of #AMLD2020 are now available on youtube. Check out the ones from the fantastic speakers we had in the #AIMolecularWorld track⬇️

@appliedmldays @befcorreia @pgainza @FreyrSverrisson Image
ML-based Design of Proteins and Small Molecules - Jennifer Listgarten (@jlistgarten)
Conditional Generation of Molecules from Disentangled Representations - Amina Mollaysa
Read 9 tweets
28 Dec 19
Looking for a weekend/holiday read?
Happy to share this major update of our #NeurIPS2019 #ML4PS workshop paper on chemical reaction classificaction (but not only.. 🧪⚗️🌍). @IBMResearch @unibern #compchem #RealTimeChem

Summary thread ⬇️:
We compared different RXN classification methods. 📍Using a BERT model borrowed from NLP, we matched the ground truth (Pistachio, @nmsoftware) with an accuracy of 98.2%. Image
We did not only visualize what was important for the class predictions by looking at the different attention weights... Image
Read 9 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(