You can use the following custom function to swap the position of two rows in a pandas DataFrame:

defswap_rows(df, row1, row2):
    df.iloc[row1], df.iloc[row2] =  df.iloc[row2].copy(), df.iloc[row1].copy()
    return df

This function will swap the positions of rows in index positions row1 and row2 in the DataFrame.

The following example shows how to use this function in practice.

Example: Swap Two Rows in Pandas

Suppose we have the following pandas DataFrame:

import pandas as pd

#create DataFrame
df = pd.DataFrame({'team' : ['Mavs', 'Nets', 'Kings', 'Cavs', 'Heat', 'Magic'],
                   'points' : [12, 15, 22, 29, 24, 22],
                   'assists': [4, 5, 10, 8, 7, 10]})

#view DataFrame
print(df)

    team  points  assists
0   Mavs      12        4
1   Nets      15        5
2  Kings      22       10
3   Cavs      29        8
4   Heat      24        7
5  Magic      22       10

We can define a swap_rows() function to swap the rows in index positions 0 and 4 in the DataFrame:

#define function to swap rows
defswap_rows(df, row1, row2):
    df.iloc[row1], df.iloc[row2] =  df.iloc[row2].copy(), df.iloc[row1].copy()
    return df

#swap rows in index positions 0 and 4
df = swap_rows(df, 0, 4)

#view updated DataFrame
print(df)

    team  points  assists
0   Heat      24        7
1   Nets      15        5
2  Kings      22       10
3   Cavs      29        8
4   Mavs      12        4
5  Magic      22       10

Notice that the rows in index positions 0 and 4 have been swapped while every other row has remained in the same position.

Note: Within the swap_rows() function, we used the function to select rows in the DataFrame based on their index position.

The following tutorials explain how to perform other common tasks in pandas: