Encode/decode a crystal structure to/from a grayscale PNG image for direct use with image-based machine learning models such as Imagen, DALLE2, or Palette.1
Star
Follow @sgbaird
Issue
Discuss
The latest advances in machine learning are often in natural language such as with long
short-term memory networks (LSTMs) and transformers or image processing such as with
generative adversarial networks (GANs), variational autoencoders (VAEs), and guided
diffusion models; however, transfering these advances to adjacent domains such as
materials informatics often takes years. xtal2png
encodes and decodes crystal
structures via grayscale PNG images by writing and reading the necessary information for
crystal reconstruction (unit cell, atomic elements, atomic coordinates) as a square
matrix of numbers, respectively. This is akin to making/reading a QR code for crystal
structures, where the xtal2png
representation is invertible. The ability to feed these
images directly into image-based pipelines allows you, as a materials informatics
practitioner, to get streamlined results for new state-of-the-art image-based machine
learning models applied to crystal structure.
Results manuscript coming soon!
:maxdepth: 2
Overview <readme>
Examples <examples>
Contributions & Help <contributing>
License <license>
Authors <authors>
Changelog <changelog>
Module Reference <api/modules>
GitHub Source <https://github.com/sparks-baird/xtal2png>
- {ref}
genindex
- {ref}
modindex
- {ref}
search
Footnotes
-
For unofficial implementations, see lucidrains/imagen-pytorch, lucidrains/DALLE2-pytorch, and Janspiry/Palette-Image-to-Image-Diffusion-Models, respectively ↩