Machine Learning by Andrew Ng---Linear Regression with multiple variables

Just as last bog,input Octave commands show in the figure below:

 

There are something different from last example with one variable,We should add a step named Feature Normalization.

using Feature Normalization will speed up your gradient descent.

To make it come true ,input commands which show in the picture below:

 

J of theta can also write like this:

so ,we can use vectorized version of J of theta like this:

also does Gradient Descent function:

Then we get theta!