Another approach to scale the data is to standardize it: transform in such a way that `mean` is equal to 0 and `std` is equal to 1. Look at the formula:

Here  μ  is a mean value of x and  σ  is a standard deviation. This way, we move each value such that total mean is 0, and normalize it with `std`, so new std is equal to 1.

Preprocessing Data is an important step in the process of analyzing data and ML problem-solving. Preparing, cleaning, and scaling data can help to make the process easier and faster. Data preprocessing is a vast area: there are various methods and approaches, and in this course, you'll learn the main ones.

Take a look at the data you are working with first.

Learn how to deal with wrong, undefined or missing values and not lose important ones.

Prepare the data for being the machine-readable and easily interpreted.

Data Standardization

解答