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

Table 2

Sensitivity tests given different maximum number of leaves in each decision tree.

Num leaves	10	100	1000	10000	100000
MSE	4.2 × 10^3	1.4 × 10^3	7.5 × 10^2	7.3 × 10^2	7.3 × 10^2
MRE	0.00255392	0.00134632	0.000719617	0.000645802	0.000645802
	2.2 × 10^−5	7.5 × 10^−6	3.9 × 10^−6	3.8 × 10^−6	3.8 × 10^−6