Pandas | 07 Function Application

To customize or other library functions used in Pandas objects, there are three important ways , let's discuss how to use these methods. Use the appropriate method depends on the function to which level ( DataFramerows or columns or elements).

Table rational function application:pipe()
Row or column function is applied:apply()
Elements function application:applymap()

First, on the whole `DataFrame`perform operations

　　By the appropriate number of parameters and functions as a pipe parameters to perform custom operations

PANDAS Import AS PD 
Import numpy AS NP 

# Adder function to add two numbers together as arguments and returns the sum 
DEF Adder (ELE1, ele2): 
   return ELE1 + ele2 

DF = pd.DataFrame (np.random.randn ( . 5 , . 3 ), = Columns [ ' col1 ' , ' col2 ' , ' col3 ' ]) 
df.pipe (Adder, 2 ) # now use a custom function to operate DataFrame 
print (df)

Output:

        col1       col2       col3
0   2.176704   2.219691   1.509360
1   2.222378   2.422167   3.953921
2   2.241096   1.135424   2.696432
3   2.355763   0.376672   1.182570
4   2.308743   2.714767   2.130288

Second,Row or column operations performed

Can be used apply()a method in DataFrameor Panelof the shaft applying any function, it is the same with the descriptive statistics, using an optional axisparameter. By default, the operation performed in columns , each column as the array.

Example -1

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])
df.apply(np.mean)
print(df)

Output:

      col1       col2        col3                                                      
0   0.343569  -1.013287    1.131245 
1   0.508922  -0.949778   -1.600569 
2  -1.182331  -0.420703   -1.725400
3   0.860265   2.069038   -0.537648
4   0.876758  -0.238051    0.473992

By passing axisparameters, can be performed on a row operation.

Example -2

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])
df.apply(np.mean,axis=1)
print(df)

输出结果：

     col1         col2         col3
0  0.543255    -1.613418    -0.500731   
1  0.976543    -1.135835    -0.719153   
2  0.184282    -0.721153    -2.876206    
3  0.447738     0.268062    -1.937888
4 -0.677673     0.177455     1.397360

Examples -3

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])
df.apply(lambda x: x.max() - x.min())
print(df)

Output:

       col1        col2      col3
0   -0.585206   -0.104938   1.424115 
1   -0.326036   -1.444798   0.196849 
2   -2.033478    1.682253   1.223152  
3   -0.107015    0.499846   0.084127
4   -1.046964   -1.935617  -0.009919

Third, perform operations on elements

Not all of the functions can be quantified (not return another array of NumPythe array, nor is any value), the DataFramemethods of applymap()and similar in the Series map()accept any Python function, and return a single value.

Example -1

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])
df['col1'].map(lambda x:x*100)
print(df)

输出结果：

       col1      col2       col3    
0    0.629348  0.088467  -1.790702 
1   -0.592595  0.184113  -1.524998
2   -0.419298  0.262369  -0.178849
3   -1.036930  1.103169   0.941882 
4   -0.573333 -0.031056   0.315590

示例-2

import pandas as pd
import numpy as np

df = pd.DataFrame(np.random.randn(5,3),columns=['col1','col2','col3'])
df.applymap(lambda x:x*100)
print(df)

输出结果：

output is as follows:
         col1         col2         col3
0   17.670426    21.969052    -49.064031
1   22.237846    42.216693     195.392124
2   24.109576   -86.457646     69.643171
3   35.576312   -162.332803   -81.743023
4   30.874333    71.476717     13.028751