Petrophysical Regression regarding Porosity, Permeability, and Water Saturation Driven by Logging-Based Ensemble and Transfer Learnings: A Case Study of Sandy-Mud Reservoirs

Figure 8

Variations of hyperparameters of KNN (a), SVR (b), and RF (c) implemented by Bayesian optimization and downtrends of RMSE of porosity values obtained by four validated predictors during the whole iteration (d). KNN = k-nearest neighbors; SVR = support vector regression; RF = random forest; LightGBM = light gradient boosting machine; RMSE = root-mean-square error.

(a)

(b)

(c)

(d)