根据数值对Pandas数据框架的行或列进行排序

在这篇文章中，我们将讨论如何在Pandas数据框架中根据数值对行或列进行排序。Pandas sort_values()方法对一个数据框架按照所传递的列的升序或降序进行排序。它与Python的排序函数不同，因为它不能对数据框进行排序，也不能选择特定的列。

语法:

DataFrame.sort_values(by, axis=0, ascending=True, inplace=False, kind=’quicksort’, na_position=’last’)

参数:

这个方法将接受以下参数。
by: 单一/列的列名来对数据框架进行排序。
axis：0或’索引’为行，1或’列’为列。
ascending: 布尔值，如果为真，则以升序排列数据框。
inplace: 布尔值。如果为真，则在传递的数据框中进行修改。
kind: 字符串，可以有三个输入(‘quicksort’, ‘mergesort’ or ‘heapsort’)用于排序数据帧的算法。
na_position:接受两个字符串输入’last’或’first’来设置Null值的位置。默认为 “最后”。

返回类型。返回一个排序的数据框，其尺寸与函数调用者的数据框相同。

现在，让我们创建一个样本数据框架。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [('Ankit', 22, 'Up', 'Geu'),
           ('Ankita', 31, 'Delhi', 'Gehu'),
           ('Rahul', 16, 'Tokyo', 'Abes'),
           ('Simran', 41, 'Delhi', 'Gehu'),
           ('Shaurya', 33, 'Delhi', 'Geu'),
           ('Harshita', 35, 'Mumbai', 'Bhu' ),
           ('Swapnil', 35, 'Mp', 'Geu'),
           ('Priya', 35, 'Uk', 'Geu'),
           ('Jeet', 35, 'Guj', 'Gehu'),
           ('Ananya', 35, 'Up', 'Bhu')
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Name', 'Age', 
                                           'Place', 'College'],
                        index =[ 'b', 'c', 'a', 'e', 'f',
                                'g', 'i', 'j', 'k', 'd'])
# show the dataframe
details

输出:

示例1：根据单列对数据框架行进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [('Ankit', 22, 'Up', 'Geu'),
           ('Ankita', 31, 'Delhi', 'Gehu'),
           ('Rahul', 16, 'Tokyo', 'Abes'),
           ('Simran', 41, 'Delhi', 'Gehu'),
           ('Shaurya', 33, 'Delhi', 'Geu'),
           ('Harshita', 35, 'Mumbai', 'Bhu' ),
           ('Swapnil', 35, 'Mp', 'Geu'),
           ('Priya', 35, 'Uk', 'Geu'),
           ('Jeet', 35, 'Guj', 'Gehu'),
           ('Ananya', 35, 'Up', 'Bhu')
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Name', 'Age',
                                           'Place', 'College'],
                        index =[ 'b', 'c', 'a', 'e', 'f',
                                'g', 'i', 'j', 'k', 'd'])
  
# Sort the rows of dataframe by 'Name' column
rslt_df = details.sort_values(by = 'Name')
  
# show the resultant Dataframe
rslt_df

输出:

示例2：根据多列对数据框架行进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [('Ankit', 22, 'Up', 'Geu'),
           ('Ananya', 31, 'Delhi', 'Gehu'),
           ('Rahul', 16, 'Tokyo', 'Abes'),
           ('Simran', 41, 'Delhi', 'Gehu'),
           ('Shaurya', 33, 'Delhi', 'Geu'),
           ('Harshita', 35, 'Mumbai', 'Bhu' ),
           ('Priya', 35, 'Mp', 'Geu'),
           ('Priya', 34, 'Uk', 'Geu'),
           ('Jeet', 35, 'Guj', 'Gehu'),
           ('Ananya', 35, 'Up', 'Bhu')
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Name', 'Age', 
                                           'Place', 'College'],
                        index =[ 'b', 'c', 'a', 'e', 'f',
                                'g', 'i', 'j', 'k', 'd'])
  
# sort Dataframe rows based on a 'Name' & 'Age' columns
  
# if duplicate value is present in 'Name' column
# then sorting will be done according to 'Age' column
rslt_df = details.sort_values(by = ['Name', 'Age'])
  
# show the resultant Dataframe
rslt_df

输出:

示例3：根据降序的列对数据框架的行进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [('Ankit', 22, 'Up', 'Geu'),
           ('Ananya', 31, 'Delhi', 'Gehu'),
           ('Rahul', 16, 'Tokyo', 'Abes'),
           ('Simran', 41, 'Delhi', 'Gehu'),
           ('Shaurya', 33, 'Delhi', 'Geu'),
           ('Harshita', 35, 'Mumbai', 'Bhu' ),
           ('Priya', 35, 'Mp', 'Geu'),
           ('Priya', 34, 'Uk', 'Geu'),
           ('Jeet', 35, 'Guj', 'Gehu'),
           ('Ananya', 35, 'Up', 'Bhu')
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Name', 'Age',
                                           'Place', 'College'],
                        index =[ 'b', 'c', 'a', 'e', 'f', 
                                'g', 'i', 'j', 'k', 'd'])
  
# sort Dataframe rows based on "Name' 
# column in Descending Order
rslt_df = details.sort_values(by = 'Name', ascending = False)
  
# show the resultant Dataframe
rslt_df

输出:

例子4：根据Place中的一列对Dataframe的行进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [('Ankit', 22, 'Up', 'Geu'),
           ('Ananya', 31, 'Delhi', 'Gehu'),
           ('Rahul', 16, 'Tokyo', 'Abes'),
           ('Simran', 41, 'Delhi', 'Gehu'),
           ('Shaurya', 33, 'Delhi', 'Geu'),
           ('Harshita', 35, 'Mumbai', 'Bhu' ),
           ('Priya', 35, 'Mp', 'Geu'),
           ('Priya', 34, 'Uk', 'Geu'),
           ('Jeet', 35, 'Guj', 'Gehu'),
           ('Ananya', 35, 'Up', 'Bhu')
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Name', 'Age',
                                           'Place', 'College'],
                        index =[ 'b', 'c', 'a', 'e', 'f',
                                'g', 'i', 'j', 'k', 'd'])
  
# Sort the rows of dataframe by  'Name' 
# column inplace
details.sort_values(by = 'Name', inplace = True)
  
# show the resultant Dataframe
details

输出:

**让我们看看另一个简单的数据框架，我们能够根据行来对列进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [
           (75, 50, 60, 70),
           (75, 55, 65, 75),
           (75, 35, 45, 25),
           (75, 90, 60, 70),
           (76, 90, 70, 60),
           (90, 80, 70, 60),
           (65, 10, 30, 20)
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Hindi', 'Math', 
                                           'Science', 'English'],
                        index = ['Ankit', 'Rahul', 'Aishwarya', 
                                 'Shivangi', 'Priya', 'Swapnil',
                                 'Shaurya'])
# show the dataframe
details

输出:

示例1：根据单一行对数据框架的列进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [
           (75, 50, 60, 70),
           (75, 55, 65, 75),
           (75, 35, 45, 25),
           (75, 90, 60, 70),
           (76, 90, 70, 60),
           (90, 80, 70, 60),
           (65, 10, 30, 20)
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Hindi', 'Math', 
                                           'Science', 'English'],
                        index = ['Ankit', 'Rahul', 'Aishwarya', 
                                 'Shivangi', 'Priya', 'Swapnil',
                                 'Shaurya'])
  
# sort columns of a Dataframe based 
# on a 'Shivangi' row
rslt_df = details.sort_values(by = 'Shivangi', axis = 1)
  
# show the dataframe
rslt_df

输出:

示例2：以降序方式对数据框架中的列进行排序，基于单一行。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [
           (75, 50, 60, 70),
           (75, 55, 65, 75),
           (75, 35, 45, 25),
           (75, 90, 60, 70),
           (76, 90, 70, 60),
           (90, 80, 70, 60),
           (65, 10, 30, 20)
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Hindi', 'Math', 
                                           'Science', 'English'],
                        index = ['Ankit', 'Rahul', 'Aishwarya', 
                                 'Shivangi', 'Priya', 'Swapnil',
                                 'Shaurya'])
  
# Sort columns of a dataframe in descending order 
# based on a 'Shivangi' row 
rslt_df = details.sort_values(by = 'Shivangi', axis = 1, ascending = False)
  
rslt_df

输出:

示例3：根据多行对数据框架的列进行排序。

# import pandas library as pd
import pandas as pd
  
# List of Tuples
students = [
           (75, 50, 60, 70),
           (75, 55, 65, 75),
           (75, 35, 45, 25),
           (75, 90, 60, 70),
           (76, 90, 70, 60),
           (90, 80, 70, 60),
           (65, 10, 30, 20)
            ]
  
# Create a DataFrame object from
# list of tuples with columns
# and indices.
details = pd.DataFrame(students, columns =['Hindi', 'Math', 
                                           'Science', 'English'],
                        index = ['Ankit', 'Rahul', 'Aishwarya', 
                                 'Shivangi', 'Priya', 'Swapnil',
                                 'Shaurya'])
  
# sort Dataframe columns based on a 'Shivangi' & 'Priya' rows
  
# if duplicate value is present in 'Shivangi'  row
# then sorting will be done according to 'Priya' row
rslt_df = details.sort_values(by = ['Shivangi', 'Priya'], axis = 1)
  
rslt_df

输出: