JSON (JavaScript Object Notion) is a handy and human-readable file format to exchange data over the internet. pandas DataFrames are the way to handle tabular data in Python. So the natural question is how to convert between the two formats.

Converting a pandas DataFrame to a JSON object is straight-forward:


The orient parameter controls how the data is formatted. index worked well for my use cases so far.

The other conversion direction is very similar:

pd.read_json(your_json_file, orient='index')

If you receive your JSON as a string in Python (e.g. through a request from an API), you must first convert the string into a JSON object for the above call to work:

import json
your_json_file = json.loads(json_as_str)