Skip to main content

Table 9 The DeepCryoPicker network architecture

From: DeepCryoPicker: fully automated deep neural network for single protein particle picking in cryo-EM

Layer Type Filters Size
I1 Input layer 227 × 227 × 3
P2 Pre-processing 227 × 227 × 3
C3 Convolution 96 11 × 11
M4 Max-pooling 3 × 3
C5 Convolution 256 5 × 5
M6 Max-pooling 3 × 3
C7 Convolution 384 3 × 3
C8 Convolution 384 3 × 3
C9 Convolution 256 3 × 3
M10 Max-pooling 3 × 3
F11 Fully connected 1 × 4096
F12 Fully connected 1 × 4096
O13 Output 1 × 1
  1. The convolutional layer and the subsampling layer are abbreviated as C and S, respectively. C3:11 × 11 × 96 means that in the third convolutional layer (C3) is comprised of 96 feature maps, each of which has a size of 11 × 11, also. C3: @27 × 27 means that output feature maps dimensions are 27 × 27 pixels