The Application of Convolutional Neural Network Combined with Fuzzy Algorithm in Colorectal Endoscopy for Tumor Assessment

Table 2

CNN model parameter settings.

Various parameters
Layers	7 layers of convolutional layer and 4 layers of pooling layer
Convolutional kernel	3 × 3
MaxPooling	2 × 2
Activation function	Multilabel classification: sigmoid neural network training model: ReLU
Optimizer	Adam
Stride	1
Dropout	25%
Learning rate	0.001
Batch size	32
Epoch	200
Number of hidden layer neurons	1024