fig2
![Borides as promising M<sub>2</sub>AX phase materials with high elastic modulus using machine learning and optimization](https://image.oaes.cc/b7ca2a16-86f4-48c7-9864-91db34874d21/jmi4017.fig.2.jpg)
Figure 2. Permutation importance plots for the top 15 features for XGB models developed using all 132 features with four random states (RS = 42, 0, 14, 28) to predict the elastic modulus for M2AX materials. These importances were used to explore stepwise addition of features for each model. XGB: XGBoost; RS: random states.