Research Article
Comparative Analysis of Skin Cancer (Benign vs. Malignant) Detection Using Convolutional Neural Networks
Table 3
Layers and parameters of ResNet50.
| | Type of layers | Outcome structure | Number of parameters |
| | Functional ResNet50 | (Nil, 7 × 7 × 2048) | 23587712 | | Conv_2 (2d) | (Nil, 5 × 5 × 64) | 1179712 | | Conv_3 (2d) | (Nil, 3 × 3 × 64) | 36928 | | Pooling (max) | (Nil, 1 × 1 × 64) | Null | | Layer flatten | (Nil, 64) | Null | | module_wrapper_8 | (Nil, 512) | 33280 | | module_wrapper_9 | (Nil, 256) | 131328 | | module_wrapper_10 | (Nil, 128) | 32896 | | module_wrapper_11 | (Nil, 64) | 8256 | | module_wrapper_12 | (Nil, 32) | 2080 | | module_wrapper_13 | (Nil, 16) | 528 | | module_wrapper_14 | (Nil, 8) | 136 | | module_wrapper_15 | (Nil, 2) | 18 | | Total number of parameters: 25,012,074; trainable parameters: | | 24,959,274; nontrainable parameters: 53,120 |
|
|
