How to Import Nested Data into a Pandas DataFrame

A quick Panda tip for you...

Import nested data into a DataFrame using json_normalize with record_path and meta.

Here’s an example 👇

import pandas as pd

data = [
    {
        "network_device": {
            "hostname": "spine1-nxos",
            "os_version": "9.3.7",
            "interface": [
                {"name": "eth0", "ip_address": "10.0.0.1"},
                {"name": "eth1", "ip_address": "10.0.0.2"},
            ],
        }
    }
]

pd.json_normalize(
    data,
    record_path=["network_device", "interface"], # list to flatten
    meta=[["network_device", "hostname"]],       # include this field with each row
)

# Output:
#    name  ip_address network_device.hostname
# 0  eth0  10.0.0.1   spine1-nxos
# 1  eth1  10.0.0.2   spine1-nxos

New to Pandas? Check out our introduction below:

An Introduction to Pandas for Network Automation

What is Pandas? Pandas is a Python library for performing data exploration, manipulation and analysis that allows you to work with data in easy-to-use Panda data structures, namely DataFrames. Pandas’ popularity can be thanked based on its ease of use, flexibility and support in helping data scientists work with large

Packet Coders - Learn Network AutomationRick Donato