BP网络是目前前馈式神经网络中应用最广泛的网络之一，实现BP算法训练神经网络完成XOR的分类问题。

设计要求：

（1）　　能够设置网络的输入节点数、隐节点数、网络层数、学习常数等各项参数；

（2）　　能够输入训练样本；

（3）　　实现BP算法的训练过程；

（4）　　实现训练过程的动态演示；

（5）　　训练完成后可输入测试数据进行测试。

-BP neural network is feedforward neural networks one of the most widely used network, training the neural network BP algorithm to achieve complete XOR classification problem. Design requirements: (a) be able to set up a network of input nodes, hidden nodes, network layers, learning constants and other parameters　(2) to enter the training samples　(3) to achieve BP algorithm training process　(4) training process to achieve dynamic presentation　(5) after completion of training input test data for testing.