Skip to content

Instantly share code, notes, and snippets.

@Nbooo

Nbooo/Keras MiniXception summary Secret

Created May 2, 2020 18:10
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/2b47b384c20e2dfb5e1224083b2ca2f3 to your computer and use it in GitHub Desktop.
Save Nbooo/2b47b384c20e2dfb5e1224083b2ca2f3 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
Keras MiniXception summary
Model: "model_1"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_1 (InputLayer) (None, 48, 48, 1) 0
__________________________________________________________________________________________________
conv2d_1 (Conv2D) (None, 46, 46, 8) 72 input_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 46, 46, 8) 32 conv2d_1[0][0]
__________________________________________________________________________________________________
activation_1 (Activation) (None, 46, 46, 8) 0 batch_normalization_1[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D) (None, 44, 44, 8) 576 activation_1[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 44, 44, 8) 32 conv2d_2[0][0]
__________________________________________________________________________________________________
activation_2 (Activation) (None, 44, 44, 8) 0 batch_normalization_2[0][0]
__________________________________________________________________________________________________
separable_conv2d_1 (SeparableCo (None, 44, 44, 16) 200 activation_2[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 44, 44, 16) 64 separable_conv2d_1[0][0]
__________________________________________________________________________________________________
activation_3 (Activation) (None, 44, 44, 16) 0 batch_normalization_4[0][0]
__________________________________________________________________________________________________
separable_conv2d_2 (SeparableCo (None, 44, 44, 16) 400 activation_3[0][0]
__________________________________________________________________________________________________
batch_normalization_5 (BatchNor (None, 44, 44, 16) 64 separable_conv2d_2[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D) (None, 22, 22, 16) 128 activation_2[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 22, 22, 16) 0 batch_normalization_5[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 22, 22, 16) 64 conv2d_3[0][0]
__________________________________________________________________________________________________
add_1 (Add) (None, 22, 22, 16) 0 max_pooling2d_1[0][0]
batch_normalization_3[0][0]
__________________________________________________________________________________________________
separable_conv2d_3 (SeparableCo (None, 22, 22, 32) 656 add_1[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 22, 22, 32) 128 separable_conv2d_3[0][0]
__________________________________________________________________________________________________
activation_4 (Activation) (None, 22, 22, 32) 0 batch_normalization_7[0][0]
__________________________________________________________________________________________________
separable_conv2d_4 (SeparableCo (None, 22, 22, 32) 1312 activation_4[0][0]
__________________________________________________________________________________________________
batch_normalization_8 (BatchNor (None, 22, 22, 32) 128 separable_conv2d_4[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D) (None, 11, 11, 32) 512 add_1[0][0]
__________________________________________________________________________________________________
max_pooling2d_2 (MaxPooling2D) (None, 11, 11, 32) 0 batch_normalization_8[0][0]
__________________________________________________________________________________________________
batch_normalization_6 (BatchNor (None, 11, 11, 32) 128 conv2d_4[0][0]
__________________________________________________________________________________________________
add_2 (Add) (None, 11, 11, 32) 0 max_pooling2d_2[0][0]
batch_normalization_6[0][0]
__________________________________________________________________________________________________
separable_conv2d_5 (SeparableCo (None, 11, 11, 64) 2336 add_2[0][0]
__________________________________________________________________________________________________
batch_normalization_10 (BatchNo (None, 11, 11, 64) 256 separable_conv2d_5[0][0]
__________________________________________________________________________________________________
activation_5 (Activation) (None, 11, 11, 64) 0 batch_normalization_10[0][0]
__________________________________________________________________________________________________
separable_conv2d_6 (SeparableCo (None, 11, 11, 64) 4672 activation_5[0][0]
__________________________________________________________________________________________________
batch_normalization_11 (BatchNo (None, 11, 11, 64) 256 separable_conv2d_6[0][0]
__________________________________________________________________________________________________
conv2d_5 (Conv2D) (None, 6, 6, 64) 2048 add_2[0][0]
__________________________________________________________________________________________________
max_pooling2d_3 (MaxPooling2D) (None, 6, 6, 64) 0 batch_normalization_11[0][0]
__________________________________________________________________________________________________
batch_normalization_9 (BatchNor (None, 6, 6, 64) 256 conv2d_5[0][0]
__________________________________________________________________________________________________
add_3 (Add) (None, 6, 6, 64) 0 max_pooling2d_3[0][0]
batch_normalization_9[0][0]
__________________________________________________________________________________________________
separable_conv2d_7 (SeparableCo (None, 6, 6, 128) 8768 add_3[0][0]
__________________________________________________________________________________________________
batch_normalization_13 (BatchNo (None, 6, 6, 128) 512 separable_conv2d_7[0][0]
__________________________________________________________________________________________________
activation_6 (Activation) (None, 6, 6, 128) 0 batch_normalization_13[0][0]
__________________________________________________________________________________________________
separable_conv2d_8 (SeparableCo (None, 6, 6, 128) 17536 activation_6[0][0]
__________________________________________________________________________________________________
batch_normalization_14 (BatchNo (None, 6, 6, 128) 512 separable_conv2d_8[0][0]
__________________________________________________________________________________________________
conv2d_6 (Conv2D) (None, 3, 3, 128) 8192 add_3[0][0]
__________________________________________________________________________________________________
max_pooling2d_4 (MaxPooling2D) (None, 3, 3, 128) 0 batch_normalization_14[0][0]
__________________________________________________________________________________________________
batch_normalization_12 (BatchNo (None, 3, 3, 128) 512 conv2d_6[0][0]
__________________________________________________________________________________________________
add_4 (Add) (None, 3, 3, 128) 0 max_pooling2d_4[0][0]
batch_normalization_12[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D) (None, 3, 3, 7) 8071 add_4[0][0]
__________________________________________________________________________________________________
global_average_pooling2d_1 (Glo (None, 7) 0 conv2d_7[0][0]
__________________________________________________________________________________________________
predictions (Activation) (None, 7) 0 global_average_pooling2d_1[0][0]
==================================================================================================
Total params: 58,423
Trainable params: 56,951
Non-trainable params: 1,472
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment