PET MAD Explorer - an exploration tool for PET-MAD features using Sketch-Map projection

The model can be downloaded directly from Hugging Face and loaded for use with PyTorch.

List of available models

PET MAD Explorer v1.0 (latest)

Downloading the Model

To download the model, you can use wget or curl to download the model checkpoint directly from the Hugging Face repository.

wget https://huggingface.co/lab-cosmo/pet-mad-explorer/resolve/main/pet-mad-explorer-latest.ckpt

Usage

You can use PET-MAD last-layer features together with a pre-trained sketch-map dimensionality reduction to obtain 2D and 3D representations of a dataset, e.g. to identify structural or chemical motifs. This can be used as a stand-alone feature builder, or combined with the chemiscope viewer to generate an interactive visualization.

import ase.io
import chemiscope
from pet_mad.explore import PETMADFeaturizer

featurizer = PETMADFeaturizer(version="latest")

# Load structures
frames = ase.io.read("dataset.xyz", ":")

# You can just compute features
features = featurizer(frames, None)

# Or create an interactive visualization in a Jupyter notebook
chemiscope.explore(
    frames,
    featurize=featurizer
)

Changelog

v1.0 - Initial PET MAD Explorer model