Land Cover Classification Using SegNet with Slope, Aspect, and Multidirectional Shaded Relief Images Derived from Digital Surface Model

Table 2

Hyper-parameter values for experiments.

Parameter	Value
Learning rate	0.001
Training cycle (iteration)	334,675
Maximum iteration	608,500
Iteration/epoch	1217/275
Validation frequency	1217
Mini-batch size	4
Shuffle	Every epoch