An Empirical Study for Adopting Machine Learning Approaches for Gas Pipeline Flow Prediction

Table 6

Details of the performance of each SVR model when using different kernel functions.

Kernel	RBF	Linear	Poly
MSE	4.6 × 10^5	6.9 × 10^5	7.1 × 10^5
MRE	2.286 × 10^−2	2.781 × 10^−2	3.614 × 10^−2
	2.457 × 10^−3	3.659 × 10^−3	3.762 × 10^−3