Copy DataFrame in Pandas

💡 Outline
You can copy DataFrame in Pandas using copy() method of DataFrame. By default, it will create deep copy of DataFrame.

Copy a DataFrame in Python

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

import pandas as pd df = pd.DataFrame([['Jay','M',21],['Jennifer','F',17],                    ['Preity','F',19],['Neil','M',17]],   By only using the assignment operator <code>=</code>, we are not creating a copy of a DataFrame. It will only create a view of the original and act as another name to access the DataFrame like a pointer.   See the following example.   <pre code = "python"> import pandas as pd df = pd.DataFrame([['Jay','M',21],['Jennifer','F',17],                    ['Preity','F',19],['Neil','M',17]],                   columns = ['Name','Gender','Age']) df1 = df df1.iloc[0,0] = 'Mark' print(df)

Output:

1 2 3 4 5 6 7

Name Gender  Age 0      Mark      M   21 1  Jennifer      F   17 2    Preity      F   19 3      Neil      M   17

As visible in the output, the changes we made to the df1 DataFrame are reflected in the original df DataFrame. To create copies of a DataFrame, we can use the copy() function provided by the pandas module.

We have two different copies available in Python. A shallow copy and a deep copy.

A shallow copy creates a reference to the original object. Any changes made to this copy are affected in the original object as well.

However, a deep copy recursively adds all the data of the original object to a new object. This way any changes made in the deep copy are not reflected in the original copy.

## Using copy() method to deep copy DataFrame in Pandas
By default, the copy function creates a deep copy. It has a deep parameter, which is set to True by default.

For example,

1 2 3 4 5 6 7 8 9 10

import pandas as pd df = pd.DataFrame([['Jay','M',21],['Jennifer','F',17],                    ['Preity','F',19],['Neil','M',17]],                   columns = ['Name','Gender','Age']) df1 = df.copy() df1.iloc[0,0] = 'Mark' print(df) print(df1)

Output:

1 2 3 4 5 6 7 8 9 10 11 12

Name Gender  Age 0       Jay      M   21 1  Jennifer      F   17 2    Preity      F   19 3      Neil      M   17        Name Gender  Age 0      Mark      M   21 1  Jennifer      F   17 2    Preity      F   19 3      Neil      M   17

In the above example, a deep copy of the original DataFrame was created. Changes made to this copy are not reflected in the original.

## Using copy() method to shallow copy DataFrame in Pandas
If we set the deep parameter as false, a shallow copy is created.

See the following code.

1 2 3 4 5 6 7 8 9 10

import pandas as pd df = pd.DataFrame([['Jay','M',21],['Jennifer','F',17],                    ['Preity','F',19],['Neil','M',17]],                   columns = ['Name','Gender','Age']) df1 = df.copy(deep = False) df1.iloc[0,0] = 'Mark' print(df) print(df1)

Output:

1 2 3 4 5 6 7 8 9 10 11 12

Name Gender  Age 0      Mark      M   21 1  Jennifer      F   17 2    Preity      F   19 3      Neil      M   17        Name Gender  Age 0      Mark      M   21 1  Jennifer      F   17 2    Preity      F   19 3      Neil      M   17

We can also use the earlier discussed assignment operator = to create shallow copies of DataFrames.

That’s all about how to copy DataFrame in Pandas.