Hands-on Deep Learning for Materials

This tool introduces users to deep learning techniques such as convolutional neural networks and variational auto encoders from a materials standpoint

Launch Tool

You must login before you can run this tool.

Version 1.1 - published on 15 Jun 2020

doi:10.21981/MRN1-6764 cite this

Open source: license | download

View All Supporting Documents



Published on


Please read the slides in the supporting docs before trying out the Jupyter notebook included in the tool!

This tool contains a self-contained deep learning demo along with a dataset. The goal of the notebook is to walk users through a typical pre-processing, modeling, and prediction workflow for convolutional neural networks and variational autoencoders using materials & chemicals data.

NB: The dataset is a small solubility dataset, and is not meant to train accurate or state-of-the-art models. The focus of this demo is on walking a user through a deep learning workflow and API (Keras/Tensorflow), in the context of materials and chemicals data.

Cite this work

Researchers should cite this work as follows:

  • Saaketh Desai, Edward Kim (2020), "Hands-on Deep Learning for Materials," https://nanohub.org/resources/citrinednn. (DOI: 10.21981/MRN1-6764).

    BibTex | EndNote