[Retracted] Prediction Model of Rotor Yarn Quality Based on CNN-LSTM

Table 4

Comparison of predicted values of two models for single intensity unevenness. Unit: %.

Serial number	Actual value	CNN-LSTM	BP	Regression algorithm
1	9.2	9.2	10	10.1
2	13.5	13.5	13	12.5
3	9.5	9.4	10	8
4	10.9	10.8	11.2	10
5	11.3	11.1	11	10.5
6	12.1	12.1	12.4	11.5
7	8.9	8.9	9.1	10
8	9.9	9.8	10.2	11
9	9.2	9.2	9	10
10	13.5	13.4	14	15
11	9.5	9.5	10.5	11
12	10.9	10.8	11.2	11.8
13	11.3	11.2	11.5	118
14	12.1	12.1	12.5	12.9
15	8.9	8.9	9.2	9.5
16	9.9	9.9	10.2	10.6
17	11.4	11.4	11.7	11.9
18	11.5	11.3	11.8	11.9
Mean absolute difference		0.055	0.450	0.566
Average relative error		0.005	0.036	0.047
Mean square error (MSE)		0.009	0.223	0.436
Root-mean-square error		0.098	0.461	0.589
square		0.990	0.880	0.550