Table 2 The details of each layer in the model
From: Semi-supervised few-shot learning approach for plant diseases recognition
Layers | Output size | Parameters | Fine-tuning |
|---|---|---|---|
Input | (84, 84, 3) | 0 | – |
Convolution | (84, 84, 64) | 1792 | Non-trainable |
Convolution | (84, 84, 64) | 36,928 | Non-trainable |
Max pooling | (42, 42, 64) | 0 | – |
Convolution | (42, 42, 128) | 73,856 | Non-trainable |
Convolution | (42, 42, 128) | 147,584 | Non-trainable |
Max pooling | (21, 21, 128) | 0 | – |
Convolution | (21, 21, 256) | 295,168 | Non-trainable |
Convolution | (21, 21, 256) | 590,080 | Non-trainable |
Convolution | (21, 21, 256) | 590,080 | Non-trainable |
Global average pool | (256) | 0 | – |
Dense | (128) | 32,896 | Trainable |
Dense | (N) | 128*N + N | Trainable |