神经网络向量化
From Ufldl
(→Forward propagation) |
|||
| Line 6: | Line 6: | ||
== Forward propagation == | == Forward propagation == | ||
| + | |||
Consider a 3 layer neural network (with one input, one hidden, and one output layer), and suppose <tt>x</tt> is a column vector containing a single training example <math>x^{(i)} \in \Re^{n}</math> . Then the forward propagation step is given by: | Consider a 3 layer neural network (with one input, one hidden, and one output layer), and suppose <tt>x</tt> is a column vector containing a single training example <math>x^{(i)} \in \Re^{n}</math> . Then the forward propagation step is given by: | ||
| - | + | ||
| - | 正向传导 | + | ==正向传导== |
| - | 我们考虑一个三层的神经网络(一个输入层、一个隐含层、以及一个输出层)，并且假定<tt>x</tt>是包含一个单一训练样本 <math>x^{(i)} \in \Re^{n}</math> 的列向量。则正向传导的步骤可以如下： | + | |
| + | 初译： 我们考虑一个三层的神经网络(一个输入层、一个隐含层、以及一个输出层)，并且假定<tt>x</tt>是包含一个单一训练样本 <math>x^{(i)} \in \Re^{n}</math> 的列向量。则正向传导的步骤可以如下： | ||
一审： | 一审： | ||
