Fig. 3
From: Machine learning enables polymer cloud-point engineering via inverse design
Machine-learning performance. (Top row) Comparison of three regression methods (a–c: linear, polynomial (order-2), and gradient boosting (with decision trees) regressions). The literature data are split into 68 training data points and 7 validation data points. Test data points are 42 experimental data points produced in the lab. The results were compared using the root-mean-squared error. We observe that GBR achieves the best generalization. (Bottom row) Final GBR model performance on three different random train-test splits of the combined data set
