Skip to content

Instantly share code, notes, and snippets.

@Nbooo
Created May 2, 2020 18:12
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save Nbooo/7d3f6f5c4dc8768f765b02408b66e6d0 to your computer and use it in GitHub Desktop.
Save Nbooo/7d3f6f5c4dc8768f765b02408b66e6d0 to your computer and use it in GitHub Desktop.
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [32, 8, 46, 46] 72
BatchNorm2d-2 [32, 8, 46, 46] 16
ReLU-3 [32, 8, 46, 46] 0
Conv2d-4 [32, 8, 44, 44] 576
BatchNorm2d-5 [32, 8, 44, 44] 16
ReLU-6 [32, 8, 44, 44] 0
Conv2d-7 [32, 8, 44, 44] 72
Conv2d-8 [32, 16, 44, 44] 128
SeparableConv2d-9 [32, 16, 44, 44] 0
BatchNorm2d-10 [32, 16, 44, 44] 32
ReLU-11 [32, 16, 44, 44] 0
Conv2d-12 [32, 16, 44, 44] 144
Conv2d-13 [32, 16, 44, 44] 256
SeparableConv2d-14 [32, 16, 44, 44] 0
BatchNorm2d-15 [32, 16, 44, 44] 32
MaxPool2d-16 [32, 16, 22, 22] 0
Conv2d-17 [32, 16, 22, 22] 128
BatchNorm2d-18 [32, 16, 22, 22] 32
Conv2d-19 [32, 16, 22, 22] 144
Conv2d-20 [32, 32, 22, 22] 512
SeparableConv2d-21 [32, 32, 22, 22] 0
BatchNorm2d-22 [32, 32, 22, 22] 64
ReLU-23 [32, 32, 22, 22] 0
Conv2d-24 [32, 32, 22, 22] 288
Conv2d-25 [32, 32, 22, 22] 1,024
SeparableConv2d-26 [32, 32, 22, 22] 0
BatchNorm2d-27 [32, 32, 22, 22] 64
MaxPool2d-28 [32, 32, 11, 11] 0
Conv2d-29 [32, 32, 11, 11] 512
BatchNorm2d-30 [32, 32, 11, 11] 64
Conv2d-31 [32, 32, 11, 11] 288
Conv2d-32 [32, 64, 11, 11] 2,048
SeparableConv2d-33 [32, 64, 11, 11] 0
BatchNorm2d-34 [32, 64, 11, 11] 128
ReLU-35 [32, 64, 11, 11] 0
Conv2d-36 [32, 64, 11, 11] 576
Conv2d-37 [32, 64, 11, 11] 4,096
SeparableConv2d-38 [32, 64, 11, 11] 0
BatchNorm2d-39 [32, 64, 11, 11] 128
MaxPool2d-40 [32, 64, 6, 6] 0
Conv2d-41 [32, 64, 6, 6] 2,048
BatchNorm2d-42 [32, 64, 6, 6] 128
Conv2d-43 [32, 64, 6, 6] 576
Conv2d-44 [32, 128, 6, 6] 8,192
SeparableConv2d-45 [32, 128, 6, 6] 0
BatchNorm2d-46 [32, 128, 6, 6] 256
ReLU-47 [32, 128, 6, 6] 0
Conv2d-48 [32, 128, 6, 6] 1,152
Conv2d-49 [32, 128, 6, 6] 16,384
SeparableConv2d-50 [32, 128, 6, 6] 0
BatchNorm2d-51 [32, 128, 6, 6] 256
MaxPool2d-52 [32, 128, 3, 3] 0
Conv2d-53 [32, 128, 3, 3] 8,192
BatchNorm2d-54 [32, 128, 3, 3] 256
Conv2d-55 [32, 7, 1, 1] 8,071
GlobalAvgPooling2d-56 [32, 7] 0
Softmax-57 [32, 7] 0
================================================================
Total params: 56,951
Trainable params: 56,951
Non-trainable params: 0
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment