Machine Learning-iv. Linear Regression with multiple Variables (Week 2)

http://blog.csdn.net/pipisorry/article/details/43529845

Machine learning machines Learning-andrew NG Courses Study notes

Multivariate linear regression multivariable linear programming

(linear regression works with multiple variables or with multiple features)

multiple Features (variables) multi-feature (variable)

{x superscript I represents the first trainning example; The x subscript I represents the first value of a particular trainning example}

The hypothesis for linear regression with multiple features (variables) representation of hypothetical functions for multivariable linear regression

additional zero feature x0 (For ease of representation)

For every example I has a feature vector X superscript i and X superscript i subscript 0 is going to being equal to 1.

Gradient descent for multiple variables multi-variable gradient descent

Model representation

The cost func minimum is solved by gradient descent algorithm to find parametersθ

{The left side is the gradient descent algorithm of the single variable linear programming solution parameters;

Right is an algorithm for solving parameters of multivariable linear programming}

Gradient descent in practice i-feature scaling gradient descent Practice 1-feature scaling

Gradient descent in practice ii-learning rate Gradient Descent Practice 2-Learning rate

Features and polynomial regression features and polynomial regression

Normal equation general equation

from:http://blog.csdn.net/pipisorry/article/details/43529845

Machine Learning-iv. Linear Regression with multiple Variables (Week 2)