How to plot json data with pandas and matplotlib?

In this step by step tutorial i will explain how to quickly create a bar and a line chart from data stored in a json (Java Script Object Notation) file using pandas and matplotlib.

Step #1: Acquire your json file

In this example we will assume that we have already acquired the json file from an API or a legacy database.

This json file text, representing actual sales vs target by area is quite simple, here are its contents:

[{
	"area": "North",
	"target": 350,
	"sales": 145
},
{
	"area": "West",
	"target": 320,
	"sales": 113
},

{
	"area": "East",
	"target": 310,
	"sales": 165
},

{
	"area": "South",
	"target": 300,
	"sales": 147
}]

Step#2: Create a DataFrame from a json

Next step is to import the pandas libray and use the pd.DataFrame constructor to initialize a DataFrame.

import pandas as pd

# define path to the json in your file system
json_file_path = 'C:\Temp\Examples\sales_targets.json'
sales = pd.read_json(json_file_path)

Note: Make sure that your Json file text is correctly formatted. If your json includes invalid characters, Python will thrwo the following ValueError exception:

ValueError: Unexpected character found when decoding array value

We now have a pandas DataFrame. Let’s look into its contents:

sales.head()

	area	target	sales
0	North	350	145
1	West	320	113
2	East	310	165
3	South	300	147

Step#3: Plot a stacked column chart with pandas

Our next step is to render a chart. Luckily pandas DataFrames has a plot() method that allows us to do exactly that:

sales.plot(x = 'area', kind='bar', stacked=True, cmap = 'viridis');

Note: we have loads of tutorials on Python Data Visualization that covers the basic of plotting with pandas and matplotlib libraries.

Here’s the stacked chart we created based on the json contents:

Step#4: Create a multiple line plot with matplotlib

Our last step is to show how we can also use matplotlib to render our chart. In this case we would like to draw a multiple line plot with matplotlib.

fig, ax = plt.subplots()
ax.plot(sales['area'], sales['target'], color = 'green', label='Targets')
ax.plot(sales['area'], sales['sales'], color = 'blue', label='Sales');
ax.set_title('Sales vs Targets');
ax.legend();

Here’s our line plot: