Prediction of hydrogen uptake of metal organic frameworks using explainable machine learning

Rapid and Accurate Screening of the COF Space for Natural Gas Purification: COFInformatics

“…The effectiveness of the ML models was assessed using several statistical accuracy metrics; the coefficient of determination ( R 2 ), mean absolute error (MAE), root-mean square error (RMSE), and Spearman’s ranking correlation coefficient (SRCC), as reported in Table S3 . Based on these metrics, as shown in Tables S4–S7 , a variety of regression models, including the Extra Tree, 55 GradientBoost, 56 and XG-Boost, 57 were selected for their ability to accurately predict CO 2 /CH 4 mixture uptakes, which will be explored in more detail in subsequent sections. All of the energetic descriptors together with other structural, chemical, and graph-based ones made available with our models for COFs in our depository .…”

Advancing CH4/H2 separation with covalent organic frameworks by combining molecular simulations and machine learning

“…The accuracies of ML models were evaluated by using the coefficient of determination ( R 2 ), mean absolute percentage error (MAPE), and root-mean square error (RMSE), which are all given in Table S3. ‡ Several regressor models as shown in Tables S4–S6 ‡ such as the Extra Tree, 53 GradientBoost, 54 XG-Boost, 55 Random Forest, 42 and LassoLarsCV 56 were selected based on their accuracies to predict CH 4 and H 2 uptakes as will be discussed in the following sections.…”