Pandas pd.pivot_table() function
Pandas common functionspd.pivot_table()is a function in the Pandas library used tocreate pivot tables. A pivot table is a common feature in spreadsheets that can group, aggregate, and reshape data, making it easy to analyze data from different perspectives.
andpd.crosstab()Compared withpivot_table(), it is more powerful, supporting multiple aggregation functions, handling missing values, and more complex data reshaping.
Word definition： pivot_tablemeans "pivot table", which is a table that can dynamically change the data layout and summarize and analyze data from multiple dimensions.
Basic syntax and parameters
pd.pivot_table()is a top-level function of the Pandas library, used to create pivot tables.
Syntax format
pd.pivot_table(data, values=None, index=None, columns=None, aggfunc='mean', fill_value=None, margins=False, dropna=True, margins_name='All')
Parameter description
- Parameter：
data- Type: DataFrame.
- Description: Source data DataFrame.
- Parameter：
values- Type: Column name or list of column names.
- Description: The column to aggregate. If not specified, all numeric columns will be aggregated.
- Parameter：
index- Type: Column name, list of column names, or function.
- Description: The column used as row index. Can be a single column or multiple columns (list); multiple columns create a hierarchical index.
- Parameter：
columns- Type: Column name or list of column names.
- Description: The column used as column index.
- Parameter：
aggfunc- Type: Function, string, or list of functions.
- Description: Aggregation function. Common values include 'sum', 'mean', 'count', 'min', 'max', 'median', 'std', etc. Default is 'mean'.
- Parameter：
fill_value- Type: Scalar or None.
- Description: The value used to fill missing values.
- Parameter：
margins- Type: Boolean.
- Description: Whether to add row and column totals. Default is
False。
Function description
- Return value: Returns a DataFrame, i.e., the pivot table.
- Effect: Reshape and summarize data according to the specified rows, columns, and aggregation function.
Examples
Let's thoroughly master the usage ofpd.pivot_table()through a series of examples from simple to complex.
Example 1: Basic usage - Create a simple pivot table
Example
import pandas as pd
import numpy as np
# 1. Create sales data
sales = pd.DataFrame({
'date': pd.date_range('2023-01-01', periods=12, freq='D'),
'product': ['A', 'B', 'C', 'A', 'B', 'C', 'A', 'B', 'C', 'A', 'B', 'C'],
'region': ['North', 'North', 'North', 'South', 'South', 'South', 'East', 'East', 'East', 'West', 'West', 'West'],
'sales': [100, 150, 200, 180, 170, 160, 190, 200, 210, 220, 230, 240]
})
print("=== Sales Data ===")
print(sales)
# 2. Create pivot table - Summarize sales by product and region
pivot = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum')
print("n=== pd.pivot_table() Sales Pivot Table ===")
print(pivot)
# 3. Use fill_value to fill missing values
print("n=== Fill missing values (fill_value=0) ===")
pivot_fill = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', fill_value=0)
print(pivot_fill)
import numpy as np
# 1. Create sales data
sales = pd.DataFrame({
'date': pd.date_range('2023-01-01', periods=12, freq='D'),
'product': ['A', 'B', 'C', 'A', 'B', 'C', 'A', 'B', 'C', 'A', 'B', 'C'],
'region': ['North', 'North', 'North', 'South', 'South', 'South', 'East', 'East', 'East', 'West', 'West', 'West'],
'sales': [100, 150, 200, 180, 170, 160, 190, 200, 210, 220, 230, 240]
})
print("=== Sales Data ===")
print(sales)
# 2. Create pivot table - Summarize sales by product and region
pivot = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum')
print("n=== pd.pivot_table() Sales Pivot Table ===")
print(pivot)
# 3. Use fill_value to fill missing values
print("n=== Fill missing values (fill_value=0) ===")
pivot_fill = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', fill_value=0)
print(pivot_fill)
Expected output:
=== 销售数据 ===
date product region sales
0 2023-01-01 A North 100
1 2023-01-01 B North 150
2 2023-01-01 C North 200
3 2023-01-01 A South 180
4 2023-01-01 B South 170
5 2023-01-01 C South 160
6 2023-01-01 A East 190
7 2023-01-01 B East 200
8 2023-01-01 C East 210
9 2023-01-01 A West 220
10 2023-01-01 B West 230
11 2023-01-01 C West 240
=== pd.pivot_table() 销售透视表 ===
region East North South West
product
A 400.0 100.0 180.0 220.0
B 400.0 150.0 170.0 230.0
C 410.0 200.0 160.0 240.0
=== 填充缺失值 (fill_value=0) ===
region East North South West
product
A 400 100 180 220
B 400 150 170 230
C 410 200 160 240
Code explanation:
- The pivot table summarizes sales by product (row) and region (column).
fill_value=0Displays missing combinations as 0 for easier reading.
Example 2: Using different aggregation functions
aggfuncThe parameter can specify multiple aggregation methods to analyze data.
Example
import pandas as pd
import numpy as np
# 1. Create richer sales data
sales = pd.DataFrame({
'product': ['A', 'B', 'A', 'B', 'A', 'B', 'A', 'B', 'A', 'B'],
'region': ['North', 'North', 'South', 'South', 'East', 'East', 'West', 'West', 'North', 'South'],
'sales': [100, 150, 200, 180, 190, 200, 220, 230, 110, 190],
'quantity': [10, 15, 20, 18, 19, 20, 22, 23, 11, 19]
})
print("=== Sales Data ===")
print(sales)
# 2. Summarize (sum)
print("n=== Sum of Sales ===")
sum_result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum')
print(sum_result)
# 3. Calculate average sales
print("n=== Average Sales ===")
mean_result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='mean')
print(mean_result.round(1))
# 4. Calculate multiple statistics at once (pass a list)
print("n=== Calculate count and mean simultaneously ===")
multi_agg = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc=['sum', 'mean'])
print(multi_agg)
import numpy as np
# 1. Create richer sales data
sales = pd.DataFrame({
'product': ['A', 'B', 'A', 'B', 'A', 'B', 'A', 'B', 'A', 'B'],
'region': ['North', 'North', 'South', 'South', 'East', 'East', 'West', 'West', 'North', 'South'],
'sales': [100, 150, 200, 180, 190, 200, 220, 230, 110, 190],
'quantity': [10, 15, 20, 18, 19, 20, 22, 23, 11, 19]
})
print("=== Sales Data ===")
print(sales)
# 2. Summarize (sum)
print("n=== Sum of Sales ===")
sum_result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum')
print(sum_result)
# 3. Calculate average sales
print("n=== Average Sales ===")
mean_result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='mean')
print(mean_result.round(1))
# 4. Calculate multiple statistics at once (pass a list)
print("n=== Calculate count and mean simultaneously ===")
multi_agg = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc=['sum', 'mean'])
print(multi_agg)
Expected output:
=== 销售数据 ===
product region sales quantity
0 A North 100 10
1 B North 150 15
2 A South 200 20
3 B South 180 18
4 A East 190 19
5 B East 200 20
6 A West 220 22
7 B West 230 23
8 A North 110 11
9 B South 190 19
=== 销售额求和 ===
region East North South West
product
A 190 210 200 220
B 200 150 370 230
=== 销售额平均值 ===
region East North South West
product
A 190 105.0 200 220
B 200 150.0 185 230
=== 同时计算数量和、均值 ===
sum mean
region East North South West East North South West
product
A 190 210 200 220 190 105.0 200 220
B 200 150 370 230 200 150.0 185.0 230
Code explanation:
- Using different aggregation functions can calculate different statistical indicators.
- When passing multiple aggregation functions at the same time, the pivot table creates multi-level column indexes.
Example 3: Using margins to add totals
Use themarginsparameter to add row and column totals, making it easier to view summary data.
Example
import pandas as pd
# Use the previous sales data
sales = pd.DataFrame({
'product': ['A', 'B', 'A', 'B', 'A', 'B'],
'region': ['North', 'North', 'South', 'South', 'East', 'East'],
'sales': [100, 150, 200, 180, 190, 200]
})
# 1. No totals
print("=== No totals ===")
print(pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum'))
# 2. Add row and column totals
print("n=== Add totals (margins=True) ===")
result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', margins=True)
print(result)
# 3. Custom total name
print("n=== Custom total name ===")
result_name = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', margins=True, margins_name='Total')
print(result_name)
# 4. Totals for multi-level row index
print("n=== Multi-level row index ===")
multi_index = pd.pivot_table(sales, values='sales', index=['product', 'region'], aggfunc='sum')
print(multi_index)
# Use the previous sales data
sales = pd.DataFrame({
'product': ['A', 'B', 'A', 'B', 'A', 'B'],
'region': ['North', 'North', 'South', 'South', 'East', 'East'],
'sales': [100, 150, 200, 180, 190, 200]
})
# 1. No totals
print("=== No totals ===")
print(pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum'))
# 2. Add row and column totals
print("n=== Add totals (margins=True) ===")
result = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', margins=True)
print(result)
# 3. Custom total name
print("n=== Custom total name ===")
result_name = pd.pivot_table(sales, values='sales', index='product', columns='region', aggfunc='sum', margins=True, margins_name='Total')
print(result_name)
# 4. Totals for multi-level row index
print("n=== Multi-level row index ===")
multi_index = pd.pivot_table(sales, values='sales', index=['product', 'region'], aggfunc='sum')
print(multi_index)
Expected output:
=== 无合计 ===
region East North South
product
A 190 100 200
B 200 150 180
=== 添加合计 (margins=True) ===
region East North South All
product
A 190 100 200 490
B 200 150 180 530
All 390 250 380 1020
=== 自定义合计名称 ===
region East North South 总计
product
A 190 100 200 490
B 200 150 180 530
总计 390 250 380 1020
=== 多层行索引 ===
sales
product region
A East 190
North 100
South 200
B East 200
North 150
South 180
Code explanation:
margins=TrueAdds "All" rows and columns, showing grand total data.margins_nameYou can customize the name of the total.
Example 4: Multi-level indexing and complex pivoting
You can use multiple columns to create a pivot table with multi-level indexing.
Example
import pandas as pd
# 1. Create data with multiple categorical variables
data = pd.DataFrame({
'year': [2022, 2022, 2022, 2023, 2023, 2023],
'quarter': ['Q1', 'Q2', 'Q3', 'Q1', 'Q2', 'Q3'],
'product': ['A', 'A', 'B', 'B', 'A', 'B'],
'sales': [100, 150, 200, 220, 180, 250]
})
print("=== Data ===")
print(data)
# 2. Use multiple columns as row index
print("n=== Multi-year multi-quarter pivot ===")
pivot = pd.pivot_table(data, values='sales', index=['year', 'quarter'], columns='product', aggfunc='sum')
print(pivot)
# 3. Multiple columns as column index
print("n=== Complex cross pivoting ===")
pivot_multi = pd.pivot_table(data, values='sales', index='year', columns=['product', 'quarter'], aggfunc='sum')
print(pivot_multi)
# 4. Use fill_value to handle missing values
print("n=== Fill missing values ===")
pivot_fill = pd.pivot_table(data, values='sales', index=['year', 'quarter'], columns='product', aggfunc='sum', fill_value=0)
print(pivot_fill)
# 1. Create data with multiple categorical variables
data = pd.DataFrame({
'year': [2022, 2022, 2022, 2023, 2023, 2023],
'quarter': ['Q1', 'Q2', 'Q3', 'Q1', 'Q2', 'Q3'],
'product': ['A', 'A', 'B', 'B', 'A', 'B'],
'sales': [100, 150, 200, 220, 180, 250]
})
print("=== Data ===")
print(data)
# 2. Use multiple columns as row index
print("n=== Multi-year multi-quarter pivot ===")
pivot = pd.pivot_table(data, values='sales', index=['year', 'quarter'], columns='product', aggfunc='sum')
print(pivot)
# 3. Multiple columns as column index
print("n=== Complex cross pivoting ===")
pivot_multi = pd.pivot_table(data, values='sales', index='year', columns=['product', 'quarter'], aggfunc='sum')
print(pivot_multi)
# 4. Use fill_value to handle missing values
print("n=== Fill missing values ===")
pivot_fill = pd.pivot_table(data, values='sales', index=['year', 'quarter'], columns='product', aggfunc='sum', fill_value=0)
print(pivot_fill)
Expected output:
=== 数据 === year quarter product sales 0 2022 Q1 A 100 1 2022 Q2 A 150 2 2022 Q3 B 200 3 2023 Q1 B 220 4 2023 Q2 A 180 5 2023 Q3 B 250 === 多年份多季度透视 === product A B year quarter 2022 Q1 100.0 NaN 2022 Q2 150.0 NaN 2022 Q3 NaN 200.0 2023 Q1 NaN 220.0 2023 Q2 180.0 NaN 2023 Q3 NaN 250.0 === 复杂交叉透视 === product A B quarter Q1 Q2 Q3 Q1 Q2 Q3 year 2022 100 150 200 NaN NaN NaN 2023 NaN 180 250 NaN NaN NaN === 填充缺失值 === product A B year quarter 2022 Q1 100 0 2022 Q2 150 0 2022 Q3 0 200 2023 Q1 0 220 2023 Q2 180 0 2023 Q3 0 250
Code explanation:
- Multi-level indexing is achieved by specifying
index=['year', 'quarter']as a list. fill_value=0Fill missing data uniformly with 0 for easier subsequent calculations.
Other extensionsTip:
pd.pivot_table()provides powerful data pivoting functionality. If you only need to count the frequency of two categorical variables,pd.crosstab()may be more concise.pd.crosstab()Returns frequency counts, whilepivot_table()can calculate multiple statistics such as average, sum, standard deviation, etc.pd.crosstab()can be regarded aspivot_table()Inaggfunc='count'a special case when