Buckle Up For Beautiful Python Plotly Mapping With GPT-4 Prompting

Author(s): John Loewen, PhD

Originally published on Towards AI.

On-the-fly interactive choropleth and bubble map code
Dall-E image: impressionist painting of mapping global undernourishment

GPT-4 prompting for Python code generation has drastically expedited my coding workflow.

I use GPT-4 with interactive libraries like Plotly and Streamlit to generate interactive data visualizations.

Recently, while working with some UN data, I was able to prompt GPT-4 to create Python Plotly code for both choropleth and scatter geo (bubble) maps to effectively visualize global undernourishment data.

I was able to combine these two maps together to provide some in-depth data storytelling about global undernourishment.

Let’s take a look together on how to combine these maps to tell this story.

My wife works for the UN Food and Agriculture Organization and is always involved with decisions related to global food security.

To find common ground for dialog and to support her work, today I am going to use a new dataset (to me). The UN food security dataset can be found HERE.

There some options to consider when downloading the dataset:

UN food security indicators for download (screenshot from UNFAO website)

For this exercise, I have chosen all of the countries for all of the years available (2000–2022), along with 4 important indicators:

Prevalence of undernourishment (%)Number of people undernourished (million)Prevalence of severe food insecurity in the total population (%)Number of… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI