How to combine two dataframe in Python – Pandas?

In some genuine circumstances, the information that we need to involve comes in numerous documents. We frequently have a need to join these records into a solitary DataFrame to break down the information. Pandas give such offices to effectively consolidating Series or DataFrame with different sorts of set rationale for the records and social polynomial math usefulness on account of join/combine type activities. Moreover, pandas likewise give utilities to think about two Series or DataFrame and sum up their disparities.

Linking DataFrames

The concat() work in pandas is utilized to add either sections or lines starting with one DataFrame then onto the next. The concat() work does all the hard work of performing link tasks along a hub while performing discretionary set rationale (association or convergence) of the files (assuming any) on different tomahawks.

import pandas as pd
# First DataFrame
df1 = pd.DataFrame({'id': ['A01', 'A02', 'A03', 'A04'],
                    'Name': ['ABC', 'PQR', 'DEF', 'GHI']})
 
# Second DataFrame
df2 = pd.DataFrame({'id': ['B05', 'B06', 'B07', 'B08'],
                    'Name': ['XYZ', 'TUV', 'MNO', 'JKL']})
 
 
frames = [df1, df2]
 
result = pd.concat(frames)
display(result)

Output:

Joining DataFrames

At the point when we linked our DataFrames we essentially added them to one another for example stacked them either in an upward direction or next to each other. One more method for consolidating DataFrames is to involve segments in each dataset that contain normal qualities (a typical special id). Consolidating DataFrames utilizing a typical field is classified “joining”. The sections containing the normal qualities are designated “join key(s)”. Joining DataFrames in this manner is regularly helpful when one DataFrame is a “query table” containing extra information that we need to remember for the other.

Note: This course of joining tables is like how we manage tables in a SQL information base.

While sticking together numerous DataFrames, you have a decision of how to deal with different tomahawks (other than the one being connected). This should be possible in the accompanying two ways :

Take the association of all, join=’outer’. This is the default choice as it brings about zero data misfortune.
Take the crossing point, join=’inner’.

Example:

import pandas as pd
 
df1 = pd.DataFrame({'id': ['A01', 'A02', 'A03', 'A04'],
                    'Name': ['ABC', 'PQR', 'DEF', 'GHI']})
 
df3 = pd.DataFrame({'City': ['MUMBAI', 'PUNE', 'MUMBAI', 'DELHI'],
                    'Age': ['12', '13', '14', '12']})
 
# the default behaviour is join='outer'
# inner join
 
result = pd.concat([df1, df3], axis=1, join='inner')
display(result)

Output:

Concatenating using append

A useful shortcut to concat() is append() instance method on Series and DataFrame. These methods actually predated concat.

Example:

import pandas as pd
# First DataFrame
df1 = pd.DataFrame({'id': ['A01', 'A02', 'A03', 'A04'],
                    'Name': ['ABC', 'PQR', 'DEF', 'GHI']})
 
# Second DataFrame
df2 = pd.DataFrame({'id': ['B05', 'B06', 'B07', 'B08'],
                    'Name': ['XYZ', 'TUV', 'MNO', 'JKL']})
 
# append method
result = df1.append(df2)
display(result)

Output:

Note: append() may take multiple objects to concatenate.

Example:

import pandas as pd
# First DataFrame
df1 = pd.DataFrame({'id': ['A01', 'A02', 'A03', 'A04'],
                    'Name': ['ABC', 'PQR', 'DEF', 'GHI']})
 
# Second DataFrame
df2 = pd.DataFrame({'id': ['B05', 'B06', 'B07', 'B08'],
                    'Name': ['XYZ', 'TUV', 'MNO', 'JKL']})
 
df3 = pd.DataFrame({'City': ['MUMBAI', 'PUNE', 'MUMBAI', 'DELHI'],
                    'Age': ['12', '13', '14', '12']})
 
 
# appending multiple DataFrame
result = df1.append([df2, df3])
display(result)

Output:

Leave a Reply

Your email address will not be published.