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

Figure 7

Variations of hyperparameters of light gradient boosting machine (LightGBM) implemented by RS (a), PSO (b), and Bayes (c), and downtrends of RMSE of porosity values generated by three applied optimizers during the whole iteration (d). RS = random research; PSO = particle swarm optimization; Bayes = Bayesian optimization; RMSE = root-mean-square error.

(a)

(b)

(c)

(d)