Sequential Human Activity Recognition Based on Deep Convolutional Network and Extreme Learning Machine Using Wearable Sensors

Table 2

CNN-LSTM architecture parameters.

Layer	Type	Number of neurons	Kernel size	Stride
1	Input	24 × 113	—	—
2	Convolutional	20 × 113	5 × 1	1
3	Convolutional	16 × 113	5 × 1	1
4	Convolutional	12 × 113	5 × 1	1
5	Convolutional	8 × 113	5 × 1	1
6‐7	LSTM	128	—	—
8	Fully connected	n	—	—