A Deep Learning Approach for Table Tennis Forehand Stroke Evaluation System Using an IMU Sensor

Table 5

RMSE, MAPE, MAE, and for the Table Tennis Forehand stroke self-collected dataset.

Strokes	Model	RMSE		MAPE	MAE
		Training	Test
Basic	RBF-SVR	4.91	4.81	0.10	8.70	0.962
	LSTM	4.79	4.44	0.07	3.30	0.998
	2DCNN	8.52	6.55	0.10	20.92	0.987
Topspin	RBF-SVR	2.45	2.51	0.05	5.12	0.965
	LSTM	4.33	2.62	0.04	2.02	0.999
	2DCNN	9.35	5.04	0.11	5.30	0.987
Push	RBF-SVR	2.08	2.20	0.04	3.80	0.961
	LSTM	3.75	2.22	0.02	2.81	0.996
	2DCNN	8.84	3.84	0.09	3.67	0.900