Fig. 3: Cross-validation root-mean-square error (RMSE) of the XGBoost method using different feature combinations.
From: Data-driven capacity estimation of commercial lithium-ion batteries from voltage relaxation
(i, j) means different feature combinations referring the Supplementary Table 10. The (7, 1) = [Var, Ske, Max] obtains the best cross-validation RMSE = 1.0% within a three feature combination.
