Replication Data for: Prediction of Electronic Density of States in Guanine-TiO2 Adsorption Model based on Machine Learning
This dataset houses the code and data related to the paper titled "Prediction of Electronic Density of States in Guanine-TiO2 Adsorption Model based on Machine Learning.” “DEFECTED_MODEL_ML” and “STOICHIOMETRIC_MODEL_ML” folders include 10 instances of neural network generations per model, whic...
| Autores: | , , |
|---|---|
| Tipo de recurso: | conjunto de datos |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Consorci de Serveis Universitaris de Catalunya (CSUC) |
| Repositorio: | CORA.Repositori de Dades de Recerca |
| OAI Identifier: | oai:dnet:cora.rdr____::8b8cb5c432be55c852b639759f3decdc |
| Acceso en línea: | https://doi.org/10.34810/DATA1223 |
| Access Level: | acceso abierto |
| Palabra clave: | Chemistry Other molecular dynamics Computational Chemistry Machine Learning Titania Guanine |
| Sumario: | This dataset houses the code and data related to the paper titled "Prediction of Electronic Density of States in Guanine-TiO2 Adsorption Model based on Machine Learning.” “DEFECTED_MODEL_ML” and “STOICHIOMETRIC_MODEL_ML” folders include 10 instances of neural network generations per model, which are numbered in the same order given in supplementary material Table S1. The “DEFECTED_MODEL_MD” and “STOICHIOMETRIC_MODEL_MD” folders provide crucial files used in our study per each time step (15050 steps) of molecular dynamics simulations. “GEOMETRIC_COORDINATES_IN_FIGURE_2” and “GEOMETRIC_COORDINATES_IN_FIGURE_3” folders provides the crucial files for each represented inset of Figure 2 and Figure 3 in the main text. Thus, one can reproduce our analysis. “MatLab_Scripts” folder provides the scripts that we used for our study. “MATLAB_ML_CVPAR_25PerCent_15Neur_2Layers” is the script for processing database. “Predict_DOS_from_GEO_URV” enables predicting DOS from Geometry. Steps are described in the code. ## Usage In example one can pick a provided figure inset folder, then can add a desired neural network and the “Predict_DOS_from_GEO_URV” script into the same folder location. Thus the predictions in the study can be reproduced. Furthermore the script enables the applications with different geometry models introduced by user. |
|---|