Table 2.

Experiment to tune hyperparameters for segmentation anything model

Parameter	Search area	Chosen value
Decay	[ 0.001, 0.0001]	0.0001
Optimizer	[Adam, RMSprop]	Adam
Learning rate	[1e-4, 2e-4]	2e-4
Dropout	[0.2, 0.1]	0.2
Batch size	[32,64,128,256]	64
Stride	[$1\times 1,2\times 2$]	$1\times 1$
Padding	[Same, Valid]	Same