Interactive NYC Taxi Data Visualization with Bokeh & Streamlit (Part 2)

Explore taxi trip patterns with dynamic visualizations, including pickup/dropoff hotspots and traffic flow, using Bokeh and Streamlit.

Animated pickup and dropoff points for NYC yellow taxi trips across Jan to June of 2016. The left chart shows pickup points from the JFK airport zone, and the right chart shows the corresponding drop off points for the trips shown on the left chart. The animation runs for 0 to 23 hours, and the color shows the trip duration (in minutes) for that hour averaged across 6 months. (Image by author)

Check out part 1 of this blog post series Kickstart Your ML Journey: Scoping, Structuring, and Exploring Data (Part 1) to understand the data and the business problem we are trying to solve.

Items covered in this post:

• Adding geo map features to the NYC taxi trip dataset
• Creating an interactive plot using Bokeh
• Deploying the bokeh plots to streamlit (https://nyc-taxi-trip-pp.streamlit.app/) — NOTE: use dark theme

Github link: Source code for this post

For this particular post, we are covering the following two topics of the business problem:

1. Identify the movement of people (in taxis) across the NY city at different hours of the day.
2. Identify the demand for taxi at different times in different parts of the city (where are the hot spots?).

As part of this, we will create an interactive visualization, with two options, which show:

1. Scatter plot: Each trip…