Construction of Sports Training Performance Prediction Model Based on a Generative Adversarial Deep Neural Network Algorithm

Table 1

Learning algorithm steps.

1	For iter < I do:
2	x ← SA(X) {sample a batch of images from a set of real images X}
3	Add a random mask to each image in the batch of images
4
5
6	Update the generative network
7	Principal component network output reconstruction dataset X
8	Else:
9	Sampling a batch of images from the reconstructed image set and the real image set
10	Update the discriminant network
11	Calculate the parameter gradient
12	Update the hash network
13	End if
14	End for