Table 2 Depiction of different parameters of ML algorithms in stage 1. Since RMSE measures the average magnitude of prediction errors, lower values indicate better model performance. Among all the models, Gaussian process regressor (GPR) demonstrates the lowest RMSE, making it the most effective model in terms of prediction accuracy. Following closely behind are support vector regression (SVR) and random forest, which also show relatively low RMSE values, indicating their strong predictive capabilities. The R square and adjusted R squared value of GPR is higher than other models. Table 3 contains the baskets of algorithms used in stage 2.

From: Machine learning enhanced ultra-high vacuum system for predicting field emission performance in graphene reinforced aluminium based metal matrix composites