Linear_reg_sklearn

这是sklearn的linear regression和前两天的statsmodel不一样，statsmodel如果用 statsmodels.api,不论是fit方法还是predict方法，都需要用sm.add_constant方法增加一列const,如果使用statsmodels.formula.api则不需要add_constant方法，只需要传入R-style formula string就可以.使用sklearn的LinearRegression则可以直接fit,predict. 只是要注意传入参数的shape.

点击(此处)折叠或打开

1. # coding: utf-8
   
2. 
   
3. import numpy as np
   
4. import statsmodels.api as sm
   
5. import seaborn as sns
   
6. import matplotlib.pyplot as plt
   
7. import pandas as pd
   
8. from sklearn.linear_model import LinearRegression
   
9. sns.set()
   
10. 
    
11. data=pd.read_csv('data-analysis/python-jupyter/1.01. Simple linear regression.csv')
    
12. 
    
13. y=data['GPA']
    
14. X=data['SAT']
    
15. print(X.shape)
    
16. print(y.shape)
    
17. 
    
18. reg=LinearRegression()
    
19. '''
    
20. run reg.fit(X,y)
    
21. error message:
    
22. ValueError: Expected 2D array, got 1D array instead:
    
23. 
    
24. check X type
    
25. type(X) == Series,
    
26. 
    
27. '''
    
28. 
    
29. #reshape
    
30. X_matrix=X.values.reshape(-1,1)
    
31. print(X_matrix.shape)
    
32. 
    
33. reg.fit(X_matrix, y)
    
34. 
    
35. '''
    
36. reg.score: R-squared
    
37. reg.coef_: coefficient / slope
    
38. reg.intercept_: intercept
    
39. '''
    
40. print(reg.score(X_matrix, y))
    
41. print(reg.coef_)
    
42. print(reg.intercept_)
    
43. 
    
44. 
    
45. #make prediction
    
46. gen_data=np.linspace(1700,1800, num=10, dtype=int)
    
47. new_data=pd.DataFrame(data=gen_data, columns=['SAT'])
    
48. reg.predict(new_data)
    
49. new_data['Predicted_GPA']=reg.predict(new_data)
    
50. print(new_data)

下面是前两天的用statsmodel.api的predict部分

点击(此处)折叠或打开

1. #predict
   
2. gen_data=np.linspace(1700,1800, num=10, dtype=int)
   
3. new_data=pd.DataFrame(data=gen_data, columns=['SAT'])
   
4. new_x=sm.add_constant(new_data)
   
5. 
   
6. predicted_y=results.predict(new_x)
   
7. 
   
8. new_x['Predicted_GPA']=predicted_y
   
9. #drop const-column
   
10. new_x=new_x.drop(['const'], axis=1)
    
11. print(new_x)