Unlike Decision Tree Classifier, some machine learning models doesn't have the ability to deal with categorical data. The categorical data are often requires a certain transformation technique if we want to include them, namely Label Encoding and One-Hot
Imbalanced datasets are a common problem in classification tasks in machine learning. Take credit card fraud prediction as a simple example: the target values are either fraud (1) or not fraud (0), but the number of fraud (1)
MSE stands for mean squared error, MAE stands for mean absolute error. These are commonly used measurement of model prediction. MSE has several advantages over MAE, but also some disadvantages.
Take the following as an example to filter a data frame based on certain conditions: There are 2 users and some of their corresponding bill dates and bill amounts, we want to only keep one record for each
To log some useful Nginx Server Configurations. Rewrite Between Webstites server { listen 80; server_name old.com www.old.com; return 301 $scheme://www.new.com$request_uri; } 301 = permanent; 302 = temporary $scheme is used to define the
Secure Shell (SSH) is a cryptographic network protocol for operating network services securely over an unsecured network. It is so far the best way to connect with your servers. GitHub also allows SSH with secure connection. Instead of
If you don't want to host your Ghost blog on a server, you may want to try Ghost Static Generator and push the content to your GitHub repo. There is one Python-based Ghost content crawler call buster. However,
Feature scaling stands for normalizing variable values into a certain range. Feature scaling can quite important for certain machine learning algorithms, such as gradient descent, support vector machine. This post is about introducing several feature scaling techniques.
