Revealed: Detailed explanation of pandas techniques for sorting by specific conditions

Pandas sorting skills revealed: How to sort according to specific conditions requires specific code examples

In the process of data processing and analysis, sorting is a very common task operate. The Pandas library is one of the powerful tools for data analysis in Python. It provides rich sorting functions that can sort data according to specific conditions. This article will introduce several commonly used sorting techniques and provide specific code examples.

1. Sort by a single column

First, let’s look at how to sort by a single column. The sort_values() function in Pandas can sort DataFrame or Series objects. Below is an example data set, we will sort by the "score" column in descending order:

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Tom', 'Jerry'],
        'score': [90, 80, 95, 85],
        'age': [25, 30, 27, 23]}

df = pd.DataFrame(data)
df_sorted = df.sort_values(by='score', ascending=False)

print(df_sorted)

Output results:

   name  score  age
2   Tom     95   27
0  Alice     90   25
3  Jerry     85   23
1    Bob     80   30

In the above code, we use sort_values()Function and set the parameter by to the column name to be sorted. In addition, ascending=False means descending sorting. If you want to sort in ascending order, set it to ascending=True.

2. Sort by multiple columns

In addition to sorting by single column, we can also sort by multiple columns. When there are multiple sorting conditions, you can use the by parameter of the sort_values() function to pass in a list containing multiple column names. The following example will be sorted in descending order according to the "score" column. If the "score" columns are the same, then sorted in ascending order according to the "age" column:

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Tom', 'Jerry'],
        'score': [90, 80, 95, 85],
        'age': [25, 30, 27, 23]}

df = pd.DataFrame(data)
df_sorted = df.sort_values(by=['score', 'age'], ascending=[False, True])

print(df_sorted)

Output result:

   name  score  age
2   Tom     95   27
0  Alice     90   25
3  Jerry     85   23
1    Bob     80   30

In the above code , we passed in a list containing two elements as the by parameter, corresponding to the two sorting conditions. At the same time, we can set the sort order of each sorting condition by passing in a list of Boolean values.

3. Sort by index

In addition to sorting by columns, we can also sort by index. The sort_index() function in Pandas can implement index sorting. Here is an example:

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Tom', 'Jerry'],
        'score': [90, 80, 95, 85],
        'age': [25, 30, 27, 23]}

df = pd.DataFrame(data)
df_sorted = df.sort_index(ascending=False)

print(df_sorted)

Output result:

   name  score  age
3  Jerry     85   23
2    Tom     95   27
1    Bob     80   30
0  Alice     90   25

In the above code, we sort the index by calling the sort_index() function. The parameter ascending=False indicates descending sorting. If you want to sort in ascending order, set it to ascending=True.

4. Custom sorting function

Sometimes, we need to sort according to a custom function. The sort_values() function in Pandas provides the parameter key, which can be passed in a function for sorting. The following is an example:

import pandas as pd

data = {'name': ['Alice', 'Bob', 'Tom', 'Jerry'],
        'score': [90, 80, 95, 85],
        'age': [25, 30, 27, 23]}

df = pd.DataFrame(data)

# 自定义排序函数，按照年龄和成绩之和进行排序
def custom_sort(row):
    return row['age'] + row['score']

df_sorted = df.sort_values(by='', key=custom_sort, ascending=False)

print(df_sorted)

Output result:

   name  score  age
2   Tom     95   27
3  Jerry     85   23
0  Alice     90   25
1    Bob     80   30

In the above code, we customized a sorting function custom_sort() and passed it insort_values()In the key parameter of the function. This function compares sizes based on the sum of the "age" and "score" columns of the input rows.

Summary:

This article introduces several aspects of Pandas sorting techniques: sorting by single column, sorting by multiple columns, sorting by index, and custom sorting functions. The flexible use of these sorting functions makes it easy to sort data according to specific conditions. I hope the sample code in this article will be helpful to everyone in practice.

The above is the detailed content of Revealed: Detailed explanation of pandas techniques for sorting by specific conditions. For more information, please follow other related articles on the PHP Chinese website!