Machine Learning

Coursera_machine_learning_assignment1 / 线性回归梯度下降

Posted by Yan on June 21, 2020

Linear Regression

m: number of samples

alpha: learning rate

Cost function: ()
- where as:
Gradient descent:
- In this case:
Code for Vectorisation:
- Cost Function: J=sum((X*theta-y).^2)/(2*m);
- Gradient Descent:
```
for iter = 1:num_iters
h=X*theta;
theta=theta-alpha/m*X'*(h-y);
end
```

Another solution for linear regression:

Normal Equations
Code: theta=pinv(X'*X)*X'*y;