5 minutes to master machine learning iris logical regression classification

This article will use 5 Minutes to help you master the most classic case of iris classification in machine learning .
sketch
Use scikit-learn library , coordination Numpy、Pandas It can make machine learning simple , Utilization based on Matplotlib Of seaborn Libraries make it easier to visualize .

First, import the library you want to use ：

from sklearn import datasets 
#  We from sklearn You can get the data in your own data set 
import numpy as np
import pandas as pd
import seaborn as sns
from sklearn.linear_model import LogisticRegression
#  Use logistic regression to learn 
from sklearn.model_selection import train_test_split 
#  Use it to segment data into training set and test set 

Import data

sklearn We have prepared some data sets for practice , Including the iris data to be used now , We just need to use datasets Of $loa{d}_{i}ris()$ The method can :

iris_data = datasets.load_iris()

Got iris_data yes sklearn Type included in , We can use $iris.keys()$ Method to see what it contains , He will return a dictionary ：

>>> iris.keys()

dict_keys(['data', 'target', 'frame',
            'target_names', 'DESCR', 'feature_names',
            'filename', 'data_module'])

It includes 150 Group data ,data Indicates the included data ,target It means label , That is, what kind of iris this flower belongs to , The iris in the data has 3 Kind of setosa, versicolor and virginica, They are contained in target_names in , Indicates the name of the label .feature_names Indicates the name of the feature , That is, the description of the characteristics of iris , For example, there are petal lengths in the data set 、 Width and calyx length 、 Width . The rest is not used in this example , Don't introduce too much .

Next, extract the data and labels , And stored in Pandas Of DataFrame in ,：

>>> data = iris.data
>>> data = data.pd.DataFrame(data,columns = iris.target_names)
#  Change the column name to the name of the feature 
>>> data.head()
	sepal length (cm)	sepal width (cm)	petal length (cm)	petal width (cm)
0	5.1	                3.5	                1.4	                0.2
1	4.9	                3.0	                1.4	                0.2
2	4.7	                3.2	                1.3	                0.2
3	4.6	                3.1	                1.5	                0.2
4	5.0	                3.6	                1.4	                0.2

Data visualization

Use seaborn Of $pairplot()$ Method can quickly view the relationship between each two variables , Including with themselves ：

sns.pairplot(data)

model

Use sklearn The estimator of builds a logistic regression model ：

modle = LogisticRegression()

Data preprocessing

First, all the data is processed into training set and test set , So that we can test the model , Use train_test_split() Method can easily do this , It will return separately x Training set ,x Test set ,y Training set ,y Test set ：

x_train,x_test,y_train,y_test = train_test_split(X=data,y=iris.target,train_size=0.8)
# 80% As a training set   The rest are used as test sets 

Training models

Using estimators $fit()$ The method can train the model ：

model.fit(x_train,y_train)

Model to evaluate

Can be directly estimated $score()$ Methods calculate the score or accuracy of the model under the test set ：

>>> model.score(x_test,y_test)
0.9333333333333333 #  The accuracy rate has reached 93.33%, This is related to the division of training set and testing machine 

Model to predict

The trained model can be used to predict the test machine data , in other words , When you know a set of data about the characteristics of iris , You can use this model to know which kind it belongs to ：

>>> s = model.predict(x_test)
array([1, 2, 1, 2, 1, 0, 2, 1, 0, 
        0, 0, 2, 1, 0, 2, 0, 1, 2,
        1, 1, 2, 2,1, 2, 0, 2, 1, 2, 0, 0])
        #  among 0 Express setosa,1 Express versicolor,2 Express virginica