Interval scaling and data number scaling

In the process of data preprocessing and algorithm implementation, it is sometimes necessary to scale related data to the same scale.

Generally speaking, scaling to the same scale mainly includes two aspects:

• The scaling of the interval (the number of data remains unchanged);
• Scaling of the number of data (change the number of data to the same number).

Two links about interval scaling:

Algorithm for mapping one interval to another

How to map a set of data from one interval to another

Scaling of the number of data:

For the scaling of the number of data, generally speaking, the main dish is the "interpolation method" (a numerical processing method)

A linear interpolation method is given here: Introduction to Linear Interpolation

There are also some libraries for linear interpolation in Python: SciPy Interpolation Library

A practical example of data scaling is given below:

Thanks for Xiaoyu.