Predicting the HOMO-LUMO Gap
of Molecules Based on Quantum Properties: Comparison of Different Learning Models.
Dataset:
Transition Metals Quantum Mechanics dataset (tmQM)
.
The dataset is extracted from Cambridge Structural Database (CSD)). It has 86,665 data points. And is publicly available at: https:// github.com/bbskjelstad/tmqm
The Neural Network model using the PyTorch framework
outperformed
the other implemented machine learning models, including Random Forest (RF), XGBoost, k-Nearest Neighbors (kNN), and Linear Regression, in terms of delivering the lowest Mean Squared Error (MSE) value.