Table 2 Summary of the architecture of the Alexnet model.
Layer (type) | Output Shape | Param # |
|---|---|---|
conv2d_1 (Conv2D) | (None, 18, 98, 96) | 23424 |
activation_1 (Activation) | (None, 18, 98, 96) | 0 |
max_pooling2d_1 (MaxPooling2 | (None, 9, 49, 96) | 0 |
batch_normalization_1 (Batch | (None, 9, 49, 96) | 384 |
xconv2d_2 (Conv2D) | (None, 1, 41, 256) | 1990912 |
activation_2 (Activation) | (None, 1, 41, 256) | 0 |
max_pooling2d_2 (MaxPooling2 | (None, 1, 21, 256) | 0 |
batch_normalization_2 (Batch | (None, 1, 21, 256) | 1024 |
conv2d_3 (Conv2D) | (None, 1, 21, 384) | 98688 |
activation_3 (Activation) | (None, 1, 21, 384) | 0 |
batch_normalization_3 (Batch | (None, 1, 21, 384) | 1536 |
conv2d_4 (Conv2D) | (None, 1, 21, 384) | 147840 |
activation_4 (Activation) | (None, 1, 21, 384) | 0 |
batch_normalization_4 (Batch | (None, 1, 21, 384) | 1536 |
conv2d_5 (Conv2D) | (None, 1, 21, 256) | 98560 |
activation_5 (Activation) | (None, 1, 21, 256) | 0 |
max_pooling2d_3 (MaxPooling2 | (None, 1, 11, 256) | 0 |
batch_normalization_5 (Batch | (None, 1, 11, 256) | 1024 |
flatten_1 (Flatten) | (None, 2816) | 0 |
dense_1 (Dense) | (None, 4096) | 11538432 |
activation_6 (Activation) | (None, 4096) | 0 |
dropout_1 (Dropout) | (None, 4096) | 0 |
batch_normalization_6 (Batch | (None, 4096) | 16384 |
dense_2 (Dense) | (None, 4096) | 16781312 |
activation_7 (Activation) | (None, 4096) | 0 |
dropout_2 (Dropout) | (None, 4096) | 0 |
batch_normalization_7 | (Batch(None, 4096) | 16384 |
dense_3 (Dense) | (None, 1000) | 4097000 |
activation_8 (Activation) | (None, 1000) | 0 |
dropout_3 (Dropout) | (None, 1000) | 0 |
batch_normalization_8 | (Batch (None, 1000) | 4000 |
dense_4 (Dense) | (None, 1) | 1001 |
activation_9 (Activation) | (None, 1) | 0 |
