Human Activity Recognition Based on a Modified Capsule Network

Table 5

Detailed structural parameters of the MCN model on the UCI-HAR dataset.

Model	Layer	Parameters
MCN	Conv2d-1	Kernel size = 3 × 3 step = 1 filters = 32
	Conv2d-2	Kernel size = 3 × 3 step = 1 filters = 64
	Conv2d-3	Kernel size = 3 × 3 step = 1 filters = 128
	Primary caps	Kernel size = 3 × 3 step = 3 filters = 32 × 8
	Activity caps	Routing iterations = 3