Table 4 Hyperparameters and model architectures.

Random Forest (RF)²⁶

Number of trees

500

Max features

Sqrt

Deep Learning (DL)²¹

Number of input parameters

12

Number of output parameters

1

Number of hidden layers

2

Number of hidden units in each layer

45, 35

Batch size

128

Learning Rate

0.001

Single-Shot MultiBox Detector (SSD)¹⁹

Big anchor shape

[(116,90), (156,198), (373,326)]

Mid anchor shape

[(30,61), (62,45), (59,119)]

Small anchor shape

[(10,13), (16,30), (33,23)]

Architecture

Darknet

Number of layers

53

Genetic-based Multi-Feature Regression (GC-MFR)²⁰

Number of records

100

Number of generations in each map

100

Number of keys

1