Table 6 Specific settings of Parameters.

From: Impact of agricultural industry transformation based on deep learning model evaluation and metaheuristic algorithms under dual carbon strategy

Parameter classification

Parameter name

Settings

CNN parameters

Number of convolutional layers

3rd floor

Convolution kernel size

The first layer: 3 × 3, the second and third layers: 5 × 5

Stride

1

Padding

‘Same’

Number of neurons

First floor: 32, second floor: 64, third floor: 128

Pooling layer

Max pooling, window size 2 × 2

LSTM parameters

Number of LSTM units

128

Learning Rate

0.001

Optimizer

Adam

Batch Size

32

Number of iterations (Epochs)

500

SMA parameters

Number of slime mold individuals

50

Food source detection threshold

0.01

Speed factor

Fast: 1.5, Slow: 0.5

Weight update strategy

Dynamic weight update