A Deep Neural Network-Based Model for Quantitative Evaluation of the Effects of Swimming Training

Table 1

Experimental results after adjusting the convolution kernel and pooling layer.

Parameter adjustment	Dropout	AUC
Optimal results	0.1763	0.7788
Convolution kernel = 1	0.1976	0.7695
Convolution kernel = 2	0.1884	0.7742
Convolution kernel = 3	0.1683	0.7681
Convolution kernel = 4	0.1739	0.7735
Convolution kernel = 5	0.197	0.7883
Convolution kernel = 6	0.1749	0.778
Convolution kernel = 7	0.1954	0.7748
Convolution kernel = 8	0.1722	0.7885