George Yu

Value Knowledge.

Linear Regression and Gradient Descent

Introduction Linear regression is the most fundamental model among all the regression models, it is simple yet powerful. In a nutshell, the goal of a linear regression model is to use a line to “f...

Nginx Server Basic Configuration

Server Blocks, TLS Requests and uWSGI Apps

OS:CentOS 7.4 Setup su -c 'yum update' or sudo yum -y update Nginx Install yum install epel-release yum install nginx systemctl start nginx systemctl enable nginx Setup Server Blocks Add a ...

A Note on the Vectorization of Multivariate Gradient Descent

The algorithm for a step of multivariate gradient descent is commonly defined as to update \[ \theta_j := \theta_j - \alpha \frac{\partial}{\partial\theta_j}J(\theta) \] for all \(j = (0\dots n)\) ...