Table 4 Parameters of the proposed CNN model for electrospray mode prediction.

From: Electrospray mode discrimination with current signal using deep convolutional neural network and class activation map

Layer

Type

Channels

Data size of each channel

Kernel size

Stride

0

Input

1

641

–

–

1

1D convolution

64

320

3

1

2

1D maxpooling

–

320

2

2

3

1D convolution

32

160

3

1

4

1D maxpooling

–

160

2

2

5

Global average pooling

32

–

–

–

6

Linear

3

–

–

–