Skip to main content

Table 1 Architecture of our proposed 1D CNN

From: Cancer type prediction based on copy number aberration and chromatin 3D structure with convolutional neural networks

Layer

Type

Output size

Conv (size, channel, pad)

Max pooling

input

in

32768*1*ch

N/A

N/A

conv1

c + r + p

8192*1*32

3*1, 32, 1

4*1

conv2

c + r + p

2048*1*64

3*1, 64, 1

4*1

conv3

c + r + p

512*1*128

3*1, 128, 1

4*1

conv4

c + r + p

128*1*256

3*1, 256, 1

4*1

conv5

c + r + p

32*1*512

3*1, 512, 1

4*1

conv6

c + r

1*1*4096

32*1, 4096, 0

N/A

fc7

fc + r + d

1*1*4096

1*1, 4096, 0

N/A

fc8

fc

1*1*25

1*1, 25, 0

N/A

loss

sm + log

1*1

N/A

N/A

  1. Annotations - in: input layer; c: convolutional layer; r: ReLU layer; p: pooling layer; fc: fully connected layer; d: dropout layer; sm: softmax layer; log: log loss layer; ch: number of input channels (depending on whether the HiC data is used); asterisk(*): multiplication