Skip to main content

Table 9 Number of parameters and shape of output in the proposed

From: MultiScale-CNN-4mCPred: a multi-scale CNN and adaptive embedding-based method for mouse genome DNA N4-methylcytosine prediction

Layers

Parameters

Output

Adaptive embedding

32

[None, 41, 8]

Bi-directional LSTM

720

[None, 41,12]

Multi-Scale CNN_3

333

[None, 39, 9]

Multi-Scale CNN_5

549

[None, 37, 9]

Multi-Scale CNN_7

765

[None, 35, 9]

Dropout_3

0

[None, 39, 9]

Dropout_5

0

[None, 37, 9]

Dropout_7

0

[None, 35, 9]

Concat

0

[None,111, 9]

Flatten

0

[None, 999]

Dense(27)

27,000

[None, 27]

Dropout

0

[None, 27]

Dense(9)

252

[None, 9]

Dense(1)

10

[None, 1]