Prompting GPT-4 For Map Creation: Testing 4 Terrific Python Libraries

Author(s): John Loewen, PhD

Originally published on Towards AI.

Automatic Python code for Folium, GeoPandas, Plotly, and Bokeh

Top highlight

Recently, GPT-4 prompting for Python code generation has improved by orders of magnitude.

This includes prompting to generate data visualizations, for example, choropleth maps.

Choropleth maps are a style of the thematic map where areas are shaded or patterned in proportion to the data point being used (i.e., value to represent relative global safety)

There are several Python libraries you can use to create choropleth maps, each with its own set of features and customization options.

For example, Folium, GeoPandas, Plotly, and Bokeh

Can GPT-4 generate working choropleth map code for all 4 of these awesome Python libraries?

The answer is — Yes!

Using a real-life development data set from the UN, let’s go through a practical working example for each of these 4 libraries.

For the purposes of testing these 4 Python libraries, we will be using “Global Peace Index” (GPI) data downloaded from the visionofhumanity.org website (located HERE).

We need to verify our data set, and we need to find a geoJSON file to combine with our dataset (to identify country boundaries).

Let’s begin by previewing the first few rows of the GPI dataset.

Prompt to GPT-4 Advanced Data Analysis (ADA): “Show me the first few rows of the dataset.”

Response from GPT-4:

Subset of the data from the GPI… 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