图4-4 改变分类符号的表示方法Change the class symbolsand the data is classified differently
图4-4所示的训练数据是我们用来训练神经网络的。
The training data shown in Figure 4-4 iswhat we use to train the neural network.
二元分类神经网络通常采用前面讨论的交叉熵函数进行训练。
The binary classification neural networkusually adopts the cross entropy function of the previous equation fortraining.
虽然不必总是采用交叉熵函数，但它对训练性能和实现过程是有益的。
You don’t have to do so all the time, butit is beneficial for the performance and implementation process.
二元分类神经网络的学习过程概括为以下步骤。
The learning process of the binaryclassification neural network is summarized in the following steps.
当然，我们使用交叉熵函数作为代价函数，使用sigmoid函数作为隐藏节点和输出节点的激活函数。
Of course, we use the cross entropyfunction as the cost function and the sigmoid function as the activation functionof the hidden and output nodes.
-
二元分类神经网络的输出层由一个输出节点组成。
-
The binary classification neural networkhas one node for the output layer.
激活函数采用sigmoid函数。
The sigmoid function is used for theactivation function.
-
使用sigmoid函数的最大值和最小值将训练数据的类别标志转换为数字表示。
-
Switch theclass titles of the training data into numbers using the maximum and minimumvalues of the sigmoid function.
-
用适当的值初始化神经网络的权值。
-
Initialize the weights of the neuralnetwork with adequate values.
-
将训练数据对{输入，正确输出}输入神经网络并获得训练输出。
-
Enter the input from the training data {input, correct output } into the neural network and obtain the output.
-
对所有训练数据重复步骤4 - 7。
-
Repeat Steps 4 - 7 for all training datapoints.
-
重复步骤4 - 8直到神经网络被正确训练。
-
Repeat Steps 4-8 until the neuralnetwork has been trained properly.
虽然以上步骤看起来很复杂，但这个过程基本上与第三章的反向传播过程相同。
Although it appears complicated because ofits many steps, this process is basically the same as that of the back-propagationof Chapter 3.
详细的解释就不再赘述了。
The detailed explanations are omitted.
——本文译自Phil Kim所著的《Matlab Deep Learning》
更多精彩文章请关注微信号：