Machine Learning Based MOF Synthesis Prediction

Paper

MOF Synthesis Prediction Enabled by Automatic Data Mining and Machine Learning
Yi Luo, Saientan Bag, Orysia Zaremba, Jacopo Andreo, Stefan Wuttke, Pascal Friederich, and Manuel Tsotsalas
Basque Center for Materials, Applications & Nanostructures
Karlsruhe Institute of Technology

Contact

• Machine learning: saientan.bag9@kit.edu, pascal.friederich@kit.edu

Requirements

• python 3.x
• Tensorflow 2
• RDKit
• Scikit-Learn

Data

• All data required to train the models is contained in this repository.
• The latest version of the full SynMOF database can be found on: https://github.com/Tsotsalas-Group/MOF_Literature_Extraction
• Contact for SynMOF database: manuel.tsotsalas@kit.edu

Machine Learning

• All python scripts can be found in the code directory
• The random forest models described in the paper can be trained using the pyhthon scripts found in the code directory
• The scripts to generate learning curves can be found in the temperature subdirectory of the RAC features

Expert quiz

• The jupyter notebook (compatible with google colab) can be found in the quiz directory
• In addition, there is a pdf printout of the expert quiz, as well as the data.zip directory required to run the quiz