Linear regression and logistic regression

This article transferred from: http://blog.csdn.net/itplus/article/details/10857843

This paper introduces in detail the linear regression and logistic regression, and introduces the principle of linear regression and the principle of logistic regression. For the logistic regression, the Newton method and quasi-Newton method can be used in the solution of the final parameter, which can be referred to as follows: http://www.cnblogs.com/ljy2013/p/5129294.html

As can be seen from the above, the standard gradient descent method uses all the training sample data to update one latitude component of a parameter at a time, and the gradient descent method is to update the component values of all the latitude of the parameter for each training sample.

Linear regression and logistic regression