Table 1 The parameter of CNN architecture.

From: Research on rolling bearing compound fault diagnosis based on AMOMCKD and convolutional neural network

Layer name

Filter numbers

Kernel size

Padding

Outputs sizes

Input

/

/

/

(2048,1)

Conv1

32

\(W_{k}(\gamma )\)

Same

(64,32)

Maxpool1

/

2

Vaild

(32,32)

Conv2

64

3

Same

(16,64)

Maxpool2

/

2

Vaild

(8,64)

FC

256

/

/

(256,1)

Softmax

/

/

/

(8,1)

Back to article page