Compared with a pig's real weight, the mean relative error of the estimated weight was 3.2 %. 实验结果表明:用该方程估算得到的种猪体重与实际称量体重的平均相对误差(MRE)为3.2%,精度较高,验证了本方法估算种猪体重的可行性。
RESULTS : The fitting precision was compared by using mean relative error and grey absolute correlation degree as evaluation criterion. 结果:以平均相对偏差和灰色绝对关联度为评价标准进行拟合精度的比较。
According to the ensemble neural network model established in this paper, the estimation result showed that the network had good stability and prediction accuracy and the estimation absolute mean relative error of 207 samples was 8.62 %. 估算结果表明,所建立的神经网络集成模型,其网络有良好的稳定性和预测精度,207个样本估算的绝对平均相对误差(MRE)为862%。
The correlation coefficient of predicted and actual values was 0.936 6, and the mean relative error was 4.44 %. These results showed a good prediction of the model. The study has guiding meaning to actual yield assessment. 模型用于对13个实验小区产量预测分析,预测值与实际值的相关系数r为0.9366,平均相对误差(MRE)为4.44%,说明模型的预测效果较好,对实际的估产工作有一定的指导意义。
Their inversion results have preferable precision, the mean absolute error is 4.36m and the mean relative error is 12 %. 反演结果具有较高精度,平均绝对误差达到4.36m,平均相对误差(MRE)达到12%。