Skip to main content
. 2022 Nov 3;10:958870. doi: 10.3389/fpubh.2022.958870

Table 8.

Comparison of experimental results with different numbers of kernel.

Number of kernels Trained accuracy Tested accuracy
Filter 1 Filter 2
16 16 0.8780 0.6732
32 0.8669 0.6753
64 0.8594 0.6832
32 32 0.8650 0.6740
64 0.8520 0.6798
128 0.8501 0.6470

The bold values are the best values obtained in the experiment.