The notes when study the Coursera class by Mr. Andrew Ng "Neural Networks & Deep Learning", section 3.3 "Computing a Neural Network's Output". It shows how to compute NN output via vectorization when there is one training example. Share it with you and hope it helps!
————————————————
Let's see how the Neural Network computes its output when there is only one training example. It's like logistic regression, but repeat a lot of times!
![](https://i-blog.csdnimg.cn/blog_migrate/71be543b8b93ed1099d279e624fc72d7.png)
Figure-1 shows how to compute the output of logistic regression.
![](https://i-blog.csdnimg.cn/blog_migrate/126d0cb114953b73e48a680f2838e494.png)
Figure-2 shows how to compute the activation units of hidden layer of NN.
Let's vectorize this:
![](https://i-blog.csdnimg.cn/blog_migrate/e316f54c79849d597d4bb3372e093503.png)
For one given training example x, we can vectorize the output of hidden layer & output layer as figure-3. Where,
<end>