用VFP实现人工神经网络的训练和运算研究

Realization of Training and Operation Research on Artificial Nerve Network in VFP

采用Visual FoxPro数据库语言对基本神经网络模型的相应算法进行编程,实现了通用神经网络参数的训练和回忆,为直接调用FoxPro信息数据和个性化神经网络的训练回忆提供了一种途径。程序测试达到了预期的效果,完成了将近450个变量的运算,是一个神经网络训练和回忆的通用系统。

To program corresponding algorithm of nerve network model using VFP Database language, the paper realizes the training and recollection of general nerve network parameter and provides one way for the training and recollection of individuality nerve network. Testing procedures achieved the desired results, completing nearly 450 variables computing,which is a neural network training and recalled common system.