Analysis and Correction of Wrong Technical Actions in Juvenile Sports Training Based on Deep Learning

Table 1

Parameter settings of DCNN.

Number of plies	Function
1	Input layer
2	Convolution layers 3-64
3	Maximum pool layer
4	Convolution layers 3-128
5	Convolution layers 3-128
6	Maximum pool layer
7	Convolution layers 3-256
8	Convolution layers 3-256
9	Maximum pool layer
10	Convolution layers 3-512
11	Convolution layers 3-512
12	Maximum pool layer
13	Convolution layers 3-512
14	Convolution layers 3-512
15	The full connection layer 3072
16	Output layer