Skip to main content
. 2018 Dec 24;19(1):59. doi: 10.3390/s19010059

Table 7.

Generalisation results for all 16 persons and both MLPs, each with 30 and 50 neurons, respectively. Here, column 1 represents the performance of our four MLPs, trained on persons 2 to 16 and tested on person 1.

Participant 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Acc.
MLP1-30 81% 50% 69% 55% 76% 59% 56% 68% 79% 68% 88% 72% 95% 72% 87% 79% 72.1%
MLP2-30 83% 51% 72% 60% 79% 62% 58% 72% 85% 74% 89% 73% 97% 75% 91% 83% 75.3%
MLP1-50 83% 49% 69% 58% 79% 62% 57% 68% 84% 70% 89% 75% 90% 74% 90% 80% 74.0%
MLP2-50 86% 52% 74% 63% 83% 65% 60% 72% 89% 74% 90% 75% 98% 75% 93% 82% 77.0%