What are ways to combine dataframes in Python?

In Python, especially with the pandas library, we have multiple methods to combine data frames. Here are some common approaches:

1. Using concat() Function

The concat() function is used to concatenate two or more data frames either vertically or horizontally. For example, if we have two data frames df1 and df2, we can merge them vertically (increasing the number of rows) as follows:

python
import pandas as pd

# Assume df1 and df2 are existing data frames
result = pd.concat([df1, df2])

To merge them horizontally (increasing the number of columns), use the axis=1 parameter:

python
result = pd.concat([df1, df2], axis=1)

2. Using merge() Function

The merge() function combines two data frames based on one or more key columns, similar to SQL JOIN operations. For example, if both data frames contain a common column "CustomerID", we can merge them on this column:

python
result = pd.merge(df1, df2, on='CustomerID')

Additionally, the merge() function allows specifying the merge type using the how parameter, which can be 'left', 'right', 'outer', or 'inner'. The default is 'inner'.

3. Using join() Function

The join() function is a simplified version of merge() for merging on indices. If the data frames' indices contain key information, we can use join() to combine them:

python
result = df1.join(df2, how='outer')

The join() function defaults to a left join, but we can specify different join types using the how parameter, such as 'left', 'right', 'inner', or 'outer'.

Example:

Suppose we have two data frames: one containing customer basic information and another containing customer purchase records. We can merge them using CustomerID to facilitate further analysis:

python
import pandas as pd

# Create example data frames
df_customers = pd.DataFrame({
    'CustomerID': [1, 2, 3],
    'Name': ['Alice', 'Bob', 'Charlie']
})

df_orders = pd.DataFrame({
    'OrderID': [101, 102, 103],
    'CustomerID': [2, 3, 1],
    'OrderAmount': [250, 150, 300]
})

# Merge data frames
result = pd.merge(df_customers, df_orders, on='CustomerID')

print(result)

This will output the merged data frame, which includes the customer ID, name, and their order information.

By using these methods, we can flexibly handle and analyze data from different sources, effectively supporting data analysis and machine learning projects.