对目标误差、网络的学习率和训练次数进行了具体的优化。
The optimum programs are made for the target error, the learning of Neural Network, the time of training.
当目标误差为0．01，径向基函数的分布常数为5时，网络达到最优化，总的正确识别率为98％。
When the error goal is 0. 01 and speed constant of radial basis function is 5, and the network achieves the optimization, and the total correct rate is 98%.
结果表明当目标误差为0．01，径向基函数的分布常数为4时，网络达到最优化，总的正确识别率为96％。
When error goal is 0. 01 and speed constant of radial basis function is 4, the network achieves optimization, and the total correct rate is 96%.
应用推荐