[Python] Sort the table by sort_values (pandas DataFrame)

[Python] How to sort the table by sort_values (DataFrame of pandas)

How to sort the DataFrame table using the "sort_values method".

There is also a sort_index method, but it is not recommended. FutureWarning: by argument to sort_index is deprecated, please use .sort_values(by=...)

1. [Basic Syntax and Main Options](# 1 Basic Syntax and Main Options)
2. [Table to use](# 2 Table to use)
3. [Sort by column](# 3 sort by column)
4. [Ascending sort](# Ascending sort)
5. [Descending Sort](#Descending Sort)
6. Sort columns with NaN (sort columns with #nan)
7. Overwrite
8. [Sort by line](# 4 sort by line)

** ▼ Basic syntax ** df.sort_values(by=['A']) └ "df": Table data └ "by = []": Column / row that is the basis for sorting * Required * └ "A": Column / row name

Creating a table

import pandas as pd
import numpy as np
​
row0 = [0, 0, 'a', 'Ah', 'one']
row1 = [2, 2, 'c', 'U', 'three']
row2 = [4, np.nan,'e', 'O', 'Samurai']
row3 = [1, 1,'b', 'I', 'Two']
row4 = [3, 3,'d', 'e', '肆']
row5 = [5, 5,'e', 'O', 'Samurai']
​
df = pd.DataFrame([row0,row1,row2,row3,row4,row5])
df.columns = ['col0', 'col1', 'col2' ,'col3', 'col4']
df.index = ['row0', 'row1', 'row2', 'row3', 'row4', 'row5']
​
df

Import numpy and use "np.nan" to use the missing value NaN.

Enter the column name that will be the sort standard in "A".

The default is "ascending = True" (optional).

Ascending sort

df.sort_values(by=['col0'])

Sorted in ascending order by the element in column col0.

You can sort alphabets and hiragana in the same way.

Sort in descending order by column name "A".

Descending sort

df.sort_values(by=['col0'], ascending=False)

・ Na_position ='last' └ Default (optional) └ Move to the bottom

・ Na_position ='first' └ Move to the top

Sort columns containing NaN (default)

df.sort_values(by=['col1'])

Sort columns containing NaN (top)

df.sort_values(by=['col1'], na_position='first')

・ ʻInplace = False` └ Default (optional) └ Do not overwrite

・ ʻInplace = True` └ Overwrite

Do not overwrite

df.sort_values(by=['col1'])
df

Overwrite

df.sort_values(by=['col1'], inplace=True)
df

The row direction can only be a representation with numbers or cannot be sorted.

• A table containing NaN or a character string will result in an error.
• An error will occur even if you specify a line that does not contain NaN or a character string.

Table to use

import pandas as pd

col0 = [10, 9, 8, 7]
col1 = [1, 10, 100, 1000]
col2 = [2, 2, 2, 2]
col3 = [0.3, 0.03, 0.003, 0.0003]
col4 = [4, 40, 400, 4000]

df2 = pd.DataFrame(col0, columns=['col0'])
df2['col1'] = col1
df2['col2'] = col2
df2['col3'] = col3
df2['col4'] = col4

df2

Ascending sort

df2.sort_values(by=[1], axis=1)

Descending sort

df2.sort_values(by=[1], axis=1, ascending=False)