Research of Epidemic Big Data Based on Improved Deep Convolutional Neural Network

Algorithm 2

The description of deep CNN based on the bagging algorithm.

Input:

The data set , where is the -th sample, the classification label is , is the number of deep CNN iterations, represents the number of network layers, is base classifiers, and the number of the base classifiers is .

Output:

Training:

Loop: for iterator to .

Step 1 (forward propagation).

For to

Step 2 (back propagation).

For to 2:

Update .

Update .

Save .

End loop.

Step 3 (the integration classification).

Input .

For

End for

Output .