Pandas: Convert a DataFrame to a List of Dictionaries

# Table of Contents

# Pandas: Convert a DataFrame to a List of Dictionaries

To convert a Pandas DataFrame to a list of dictionaries:

Call the DataFrame.to_dict() method on the DataFrame.
Set the orient argument to "records" when calling to_dict().
The to_dict() method will convert the DataFrame to a list of dictionaries.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = df.to_dict('records')

# [{'name': 'Alice', 'experience': 1, 'salary': 175.1}, {'name': 'Bobby', 'experience': 3, 'salary': 180.2}, {'name': 'Carl', 'experience': 5, 'salary': 190.3}, {'name': 'Dan', 'experience': 7, 'salary': 205.4}]
print(list_of_dicts)

convert dataframe to list of dictionaries

The code for this article is available on GitHub

By default, the DataFrame.to_dict() method converts the DataFrame to a dictionary.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

a_dict = df.to_dict()

# {'name': {0: 'Alice', 1: 'Bobby', 2: 'Carl', 3: 'Dan'}, 'experience': {0: 1, 1: 3, 2: 5, 3: 7}, 'salary': {0: 175.1, 1: 180.2, 2: 190.3, 3: 205.4}}
print(a_dict)

by default to dict method converts dataframe to dictionary

However, when the orient argument is set to "records", the to_dict method converts the DataFrame to a list of dictionaries.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = df.to_dict('records')

print(list_of_dicts)

print(type(list_of_dicts))

print(type(list_of_dicts[0]))

Running the code sample produces the following output.

shell

Copied!
[{'name': 'Alice', 'experience': 1, 'salary': 175.1}, {'name': 'Bobby', 'experience': 3, 'salary': 180.2}, {'name': 'Carl', 'experience': 5, 'salary': 190.3}, {'name': 'Dan', 'experience': 7, 'salary': 205.4}]

<class 'list'>

<class 'dict'>

convert dataframe to list of dictionaries using orient records

The type class returns the type of an object.

# Including the index value in each dictionary in the list

If you need to include the index value of each row in the dictionaries, call the DataFrame.reset_index() method before calling to_dict().

shell

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = df.reset_index().to_dict('records')

# [{'index': 0, 'name': 'Alice', 'experience': 1, 'salary': 175.1}, {'index': 1, 'name': 'Bobby', 'experience': 3, 'salary': 180.2}, {'index': 2, 'name': 'Carl', 'experience': 5, 'salary': 190.3}, {'index': 3, 'name': 'Dan', 'experience': 7, 'salary': 205.4}]
print(list_of_dicts)

include index value in each dictionary in the list

If you want to have the indices as dictionary keys, set the orient argument to "index" when calling to_dict().

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = df.to_dict('index')

# {0: {'name': 'Alice', 'experience': 1, 'salary': 175.1}, 1: {'name': 'Bobby', 'experience': 3, 'salary': 180.2}, 2: {'name': 'Carl', 'experience': 5, 'salary': 190.3}, 3: {'name': 'Dan', 'experience': 7, 'salary': 205.4}}
print(list_of_dicts)

setting orient argument to index

The code for this article is available on GitHub

When the orient argument is set to "index", the to_dict() method returns a nested dictionary where the indices are the dictionary's keys.

# Pandas: Convert a DataFrame to a List of Dictionaries using `values()`

You can also use the dict.values() method after transposing the DataFrame to convert it to a list of dictionaries.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = list(df.T.to_dict().values())

# [{'name': 'Alice', 'experience': 1, 'salary': 175.1}, {'name': 'Bobby', 'experience': 3, 'salary': 180.2}, {'name': 'Carl', 'experience': 5, 'salary': 190.3}, {'name': 'Dan', 'experience': 7, 'salary': 205.4}]
print(list_of_dicts)

convert dataframe to list of dictionaries using values

The DataFrame.T property is an accessor for the DataFrame.transpose method.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

#                 0      1      2      3
# name        Alice  Bobby   Carl    Dan
# experience      1      3      5      7
# salary      175.1  180.2  190.3  205.4
print(df.T)

The code for this article is available on GitHub

The transpose() method writes the DataFrame rows as columns and vice versa.

Once you convert the transposed DataFrame to a dictionary with to_dict(), call the values() method on the result.

main.py

Copied!
list_of_dicts = list(df.T.to_dict().values())

The dict.values() method returns a view of the dictionary's values.

The method returns a view object, so we had to use the list class to convert the result to a list.

# Only converting a specific column of the `DataFrame` to a list of dictionaries

If you only need to convert a specific column of the DataFrame to a list of dictionaries, use two sets of square brackets to select the column and call the to_dict() method on the result.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

list_of_dicts = df[['experience']].to_dict('records')

# [{'experience': 1}, {'experience': 3}, {'experience': 5}, {'experience': 7}]
print(list_of_dicts)

only convert specific column of dataframe to list of dictionaries

The code for this article is available on GitHub

Notice that we used two sets of square brackets [[]] when selecting the "experience" column.

main.py

Copied!
import pandas as pd


df = pd.DataFrame({
    'name': ['Alice', 'Bobby', 'Carl', 'Dan'],
    'experience': [1, 3, 5, 7],
    'salary': [175.1, 180.2, 190.3, 205.4],
})

#    experience
# 0           1
# 1           3
# 2           5
# 3           7
print(df[['experience']])

The last step is to call the to_dict() method on the result, setting the orient argument to "records".

main.py

Copied!
list_of_dicts = df[['experience']].to_dict('records')

When the orient argument is set to "records", the to_dict method returns a list of dictionaries containing the results.

# Additional Resources

You can learn more about the related topics by checking out the following tutorials:

I wrote a book in which I share everything I know about how to become a better, more efficient programmer.

You can use the search field on my Home Page to filter through all of my articles.

Pandas: Convert a DataFrame to a List of Dictionaries

# Table of Contents

# Pandas: Convert a DataFrame to a List of Dictionaries

# Including the index value in each dictionary in the list

# Pandas: Convert a DataFrame to a List of Dictionaries using values()

# Only converting a specific column of the DataFrame to a list of dictionaries

# Additional Resources

Borislav Hadzhiev

# Pandas: Convert a DataFrame to a List of Dictionaries using `values()`

# Only converting a specific column of the `DataFrame` to a list of dictionaries