Quantitative Prediction of Low-Permeability Sandstone Grain Size Based on Conventional Logging Data by Deep Neural Network-Based BP Algorithm

Table 4

Analysis of the training results of the MD machine learning method for well X1.

Method	Metrics for forecasting
	RMSE	MAE
Multiple regression analysis fitting	104.7416	70.59334	0.5554
Decision tree	65.15264	35.43207	0.8203
SVM	66.36261	29.83985	0.8843
Single hidden layer 1000 epochs	27.56469	10.64478	0.9693
Single hidden layer 2000 epochs	14.64674	7.859871	0.9722
Three hidden layer 1000 epochs	8.239049	2.951308	0.9792
Three hidden layer 2000 epochs	1.340285	0.688508	0.9831