/MOFSNN

A model to predict mutiple stability labels for MOFs.

Primary LanguageJupyter Notebook

MOF Stability prediction Neural Network (MOFSNN)

This is a study of predicting various types of MOF stability using multi-task strategy.

Raw data

The raw data is stored in the raw_data folder, which can get from literature.

TSD & SSD

The thermal stability dataset and solvent stability dataset are get from Nandy's work. The data is download from here, which is published in Nandy's another paper. The CIF files are matched with MOFs in the CoREMOF2019 databse, which is available from MOFX-DB.

WS24

The water stability, acid stability, base stability, and boiling stability dataset are get from Terrones's work. And the raw data is download from here.

Download and extract data

wget https://zenodo.org/records/5737968/files/SciData.zip?download=1 -O Nandy_2022_SciData.zip
wget https://zenodo.org/api/records/12110918/files-archive -O WS24v2.zip
wget https://mof.tech.northwestern.edu/Datasets/CoREMOF%202019-mofdb-version:dc8a0295db.zip -O CoREMOF2019.zip
unzip Nandy_2022_SciData.zip
unzip WS24v2.zip
unzip CoREMOF2019.zip