Sklearn Library -- linear regression model

Catalog

Simple linear regression

Multiple linear regression

Classification variables are converted into dummy variables

---

Yes statsmodels The library can be collated with sklearn The collation of the library can be used for reference

statsmodels library —— linear regression model _ I am a little monster blog -CSDN Blog  

Simple linear regression

from sklearn import linear_model# Import the required package 
lr=linear_model.LinearRegression()# Specify the model 
# Again fit Specify arguments and response variables in , Note that capitalization is used here X And lowercase y
# because sklearn It is accepted that numpy Array , Therefore, it is necessary to process data , In order to dataframe Pass in sklearn.
predicted=lr.fit(X=tips['total_bill'].values.reshape(-1,1),y=tips['tip'])
print(predicted.coef_)# By fitting the model coef_ Property to get the coefficient 
print('----'*6)# Output horizontal lines to distinguish each output 
print(predicted.intercept_)# By fitting the model intercept_ Property to get the intercept 
print('----'*6)
y_pre=lr.predict(tips['total_bill'][0:2].values.reshape(-1,1))# The predicted value is obtained according to the fitted linear model 
print(y_pre)# Output predicted value 
print(tips['tip'][0:2])# Output the actual value for comparison 

[0.10502452]
------------------------
0.9202696135546731
------------------------
[2.70463616 2.00622312]
0    1.01
1    1.66
Name: tip, dtype: float64

Multiple linear regression

from sklearn import linear_model# Import the required package 
lr=linear_model.LinearRegression()
predicted=lr.fit(X=tips[['total_bill','size']],y=tips['tip'])# When multiple columns are passed in, the list is used to pass in 
print(predicted.coef_)
print('----'*6)
print(predicted.intercept_)

[0.09271334 0.19259779]
------------------------
0.6689447408125031

Classification variables are converted into dummy variables

Because of sklearn You need to manually create virtual variables when you encounter classification variables , Use pandas Medium get_dummies function

import pandas as pd
# Create virtual variables , among drop_first Parameter specifies whether to delete the reference variable 
tips_dummy=pd.get_dummies(tips[['total_bill','size','sex','smoker','day','time']],drop_first=False)
print(tips_dummy.head())

   total_bill  size  sex_Male  sex_Female  smoker_Yes  smoker_No  day_Thur  \
0       16.99     2         0           1           0          1         0   
1       10.34     3         1           0           0          1         0   
2       21.01     3         1           0           0          1         0   
3       23.68     2         1           0           0          1         0   
4       24.59     4         0           1           0          1         0   

   day_Fri  day_Sat  day_Sun  time_Lunch  time_Dinner  
0        0        0        1           0            1  
1        0        0        1           0            1  
2        0        0        1           0            1  
3        0        0        1           0            1  
4        0        0        1           0            1  

from sklearn import linear_model# Import the required package 
import pandas as pd
# Create virtual variables , among drop_first Parameter specifies whether to delete the reference variable , For example, gender is divided into male and female , Then the system will select the first male as the reference variable , After deletion, the male column will not be converted into a dummy variable 
tips_dummy_drop=pd.get_dummies(tips[['total_bill','size','sex','smoker','day','time']],drop_first=True)
print(tips_dummy_drop.head())
lr=linear_model.LinearRegression()
predicted=lr.fit(X=tips_dummy_drop,y=tips['tip'])# When multiple columns are passed in, the list is used to pass in 
print(predicted.coef_)
print(predicted.intercept_)

   total_bill  size  sex_Female  smoker_No  day_Fri  day_Sat  day_Sun  \
0       16.99     2           1          1        0        0        1   
1       10.34     3           0          1        0        0        1   
2       21.01     3           0          1        0        0        1   
3       23.68     2           0          1        0        0        1   
4       24.59     4           1          1        0        0        1   

   time_Dinner  
0            1  
1            1  
2            1  
3            1  
4            1  
------------------------
[ 0.09448701  0.175992    0.03244094  0.08640832  0.1622592   0.04080082
  0.13677854 -0.0681286 ]
------------------------
0.590837425951376