Table 1 Neural network model and parameters used for the present study (simple model).

From: Deep learning and taphonomy: high accuracy in the classification of cut marks made on fleshed and defleshed bones using convolutional neural networks

Layer (type)

Output Shape

Param #

conv2d_1 (Conv2D)

(None, 78, 398, 32)

896

max_pooling2d_1(MaxPooling2

(None, 39, 199, 32)

0

conv2d_2 (Conv2D)

(None, 37, 197, 64)

18496

max_pooling2d_2 (MaxPooling2

(None, 18, 98, 64)

0

conv2d_3 (Conv2D)

(None, 16, 96, 128)

73856

max_pooling2d_3 (MaxPooling2

(None, 8, 48, 128)

0

conv2d_4 (Conv2D)

(None, 6, 46, 128)

147584

max_pooling2d_4 (MaxPooling2

(None, 3, 23, 128)

0

flatten_1 (Flatten)

(None, 8832)

0

dense_1 (Dense)

(None, 512)

4522496

dense_2 (Dense)

(None, 1)

513

  1. Total params: 4,763,841.
  2. Trainable params: 4,763,841.
  3. Non-trainable params: 0.
Back to article page