| Autor: |
Gvozdetskyi, Volodymyr, Selvaratnam, Balaranjan, Oliynyk, Anton O., Mar, Arthur |
| Rok vydání: |
2022 |
| DOI: |
10.6084/m9.figshare.19895788.v3 |
| Popis: |
This folder contains the Jupyter notebook files (and PDF snapshots) used for the ML models and the SISSO calculation files. 1. RX_data_ml_with_split.csv - Dataset file with features, labels, and indicator for train-test splits. 2. Figure_5_plot_data.csv - predicted class and probabilities by the ML models. 3. RX_Classification_final_v2 - Optimization of SVC, RF, and DT models. 4. RX_Data_for_Table_and_SI - Calculation of feature importance and average metrics for SVC, RF, and DT models. 5. SISSO_Regression - Folder containing SISSO files for the regression to find the y axis expression. 6. SISSO_Classification - Folder containing SISSO classification for the RX data set. This calculation used same features used ofr other ML models for SISSO classification. 7. SISSO_Classification_Figure_5 - SISSO classification calculation to separate 'puzzle' group from others. Description dataset System - Elements in the equiatomic binary intermetallic compound Structure_type - True structure type label - Numerical value for the structure type (for machine learning) A - Rare earth element B - X element A_Z - Atomic number of rare earth element A_R_m - Metallic radii of rare earth element (Angstroms) B_Z - Atomic number of X element B_EN_p - Pauling electronegativity of X element B_Ep - Period of X element B_Eg - Group of X element B_R_co - Covalent radii of X element (Angstroms) split - Indicator variable for train/test splits |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
