If True, then the source DataFrame will be directly modified, and no new DataFrame will be created.
If False, then a new DataFrame will be created, and the source DataFrame will be kept intact.

By default, inplace=False.

5. kind | string | optional

The sorting algorithm to use:

Kind	Speed	Worst case	Memory	Stable
quicksort	1 (fast)	`O(n^2)`	0	no
mergesort	2	`O(nlogn)`	~n/2	yes
heapsort	3 (slow)	`O(nlogn)`	0	no

By default, kind="quicksort".

NOTE

Sorting algorithms that are "stable" retain the relative ordering of duplicate values. For instance, suppose you are sorting the array [(2,3), (2,1), (4,5)] by the first element of each tuple. We have a duplicate value of 2 here, and stable sorting algorithms ensure that (2,3) will always come before (2,1) since that is how they are ordered originally. Unstable searches provide no guarantee that such ordering is retained.

6. na_positionlink | string | optional

Where to place NaN values:

Value	Description
`"first"`	Place `NaN`s at the beginning
`"last"`	Place `NaN`s at the end

By default, na_position="last".

7. ignore_indexlink | boolean | optional

If True, then the index of the sorted DataFrame will be 0,1,...,n-1, where n is the number of rows of the DataFrame.
If False, then the index names will be kept as is.

By default, ignore_index=False.

Return Value

A DataFrame sorted by row or column values.

Examples

Consider the following DataFrame:


        
        
            
                
                
                    df = pd.DataFrame({"A":[5,3,1,3],"B":["c","d","a","a"]})
df
                
            
               A  B
0  5  c
1  3  d
2  1  a
3  3  a

Sorting by column value

To sort by column A:


        
        
            
                
                
                    df.sort_values("A")
                
            
               A  B
2  1  a
1  3  d
3  3  a
0  5  c

Sorting by multiple columns

To sort by multiple columns, pass in a list of column labels:


        
        
            
                
                
                    df.sort_values(["A","B"])
                
            
               A  B
2  1  a
3  3  a
1  3  d
0  5  c

Here, the rows are sorted first by the values in column A. When there are duplicate values, we sort by the values in column B. This is the reason why the value "a" is guaranteed to come before "d" in this case.

Sorting by row value

Consider the following DataFrame:


        
        
            
                
                
                    df = pd.DataFrame({"A":[5,9,4],"B":[1,3,2]}, index=["a","b","c"])
df
                
            
               A  B
a  5  1 
b  9  3
c  4  2

To sort by values in row "b":


        
        
            
                
                
                    df.sort_values("b", axis=1)
                
            
               B  A
a  1  5
b  3  9
c  2  4

Sorting in descending order

Consider the following DataFrame:


        
        
            
                
                
                    df = pd.DataFrame({"A":[5,9,5],"B":[4,3,1]}, index=["a","b","c"])
df 
                
            
               A  B
a  5  4
b  9  3
c  5  1

By default, values are sorted in ascending order. To sort in descending order instead, set ascending=False:


        
        
            
                
                
                    df.sort_values("A", ascending=False)
                
            
               A  B
b  9  3
a  5  4
c  5  1

Passing an array of booleans

If you're sorting by multiple columns, you can pass a list of booleans to specify how each column should be sorted.


        
        
            
                
                
                    df.sort_values(["A","B"], ascending=[False, True])
                
            
               A  B
b  9  3
c  5  1
a  5  4

Here, we are sorting by column A values in descending order, while sorting by column B in ascending order.

Specifying na_position

Consider the following DataFrame:


        
        
            
                
                
                    df = pd.DataFrame({"A":[pd.np.NaN,5,4]}, index=["a","b","c"])
df
                
            
               A
a  NaN
b  5.0
c  4.0

By default, na_position="last", which means that NaN are placed at the end:


        
        
            
                
                
                    df.sort_values(by="A")   # na_position="last"
                
            
               A
c  4.0
b  5.0
a  NaN

To place NaN in the beginning instead:


        
        
            
                
                
                    df.sort_values(by="A", na_position="first")
                
            
               A
a  NaN
c  4.0
b  5.0

Specifying ignore_index

Consider the following DataFrame:


        
        
            
                
                
                    df = pd.DataFrame({"A":[5,9,2]}, index=["a","b","c"])
df
                
            
               A
a  5
b  9
c  2

By default, the index names are kept:


        
        
            
                
                
                    df.sort_values(by="A")
                
            
               A
c  2
a  5
b  9

By setting ignore_index=True, we can reset the index:


        
        
            
                
                
                    df.sort_values(by="A", ignore_index=True)
                
            
               A
0  2
1  5
2  9

Published by Isshin Inada

Edited by 0 others

Did you find this page useful?

thumb_up

thumb_down

Comment

Citation

Ask a question or leave a feedback...

Official Pandas Documentation

https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.sort_values.html

thumb_up

thumb_down

chat_bubble_outline

settings

Enjoy our search

Hit / to insta-search docs and recipes!