Shulou(Shulou.com)06/02 Report--

Today, I will talk to you about how to achieve data scaling in sklearn, many people may not know much about it. In order to make you understand better, the editor has summarized the following content for you. I hope you can get something according to this article.

I. data scaling

Your preprocessed dataset may contain a variety of feature fields with different proportions, such as US dollars, kilograms, and prices. Some feature columns range from 100 to millions.

Many machine learning models do not want such a big difference, if the smaller the numerical difference between the attributes of the data set, the better the model will perform. Here is not a generalization, the specific reasons for friends to explore their own.

Method 1: data normalization

Data normalization refers to scaling the original data to between 0 and 1.

This scaling of input attributes works well for models that depend on the size of the value. For example, the distance measurement in K-nearest neighbor model and the preparation of regression coefficient.

Next, demonstrate data normalization with the well-known iris dataset:

# normalize the iris data set.

From sklearn.datasets import load_iris

From sklearn import preprocessing

# load data

Iris = load_iris ()

Print (iris.data.shape)

# separate the original data set, divided into independent variables and dependent variables

X = iris.data

Y = iris.target

# normalization processing

Normalized_X = preprocessing.normalize (X)

Method 2: data standardization

Data standardization refers to scaling the data and changing the distribution of each attribute so that the average value is 0 and the standard deviation is 1.

For models that depend on feature distribution, such as Gaussian processes, it is very useful to standardize features.

The example of still using irises:

# standardize the data set of Iris.

From sklearn.datasets import load_iris

From sklearn import preprocessing

# load data

Iris = load_iris ()

Print (iris.data.shape)

# separate the original data set, divided into independent variables and dependent variables

X = iris.data

Y = iris.target

# standardized processing

Standardized_X = preprocessing.scale (X)

After reading the above, do you have any further understanding of how to achieve data scaling in sklearn? If you want to know more knowledge or related content, please follow the industry information channel, thank you for your support.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.