Next Steps

Congratulations on completing the practical! Here are some useful resources.

Stack Overflow & Forums

These communities are invaluable:

For R

For Python

Modern High-Performance Tools

  • DuckDB - Fast, embedded SQL analytical engine, great for querying large CSV, Parquet, and transport datasets directly
  • Polars - Lightning-fast multi-threaded DataFrame library for Python and Rust

Statistics with R/Python:

Build Your Portfolio

Project Ideas

  1. Analyse local transport patterns using open transport data
  2. Recreate published analyses to learn techniques
  3. Explore open road safety data using UK crash statistics
  4. Contribute to open source R or Python packages

Share Your Work

  • Create a GitHub profile for your code (use the Student Developer Pack for free perks!)
    • Note: Keep university assessment data private unless authorised.
  • Build a portfolio website with Quarto
  • Write blog posts about what you learn
  • Share projects on social media (#RStats, #DataScience)

Go Further: A Five-Week Course

If this practical clicked and you want the long version, Python for Transport & Civil Engineering by Dr Chris Rushton at the Institute for Transport Studies here at Leeds is a free, five-week course aimed at transport and civil engineering students. It starts where we started — running your first script, reading a traceback — and ends with you building a transport atlas of a place you choose from national open data (NaPTAN stops, STATS19 road casualties, IMD, Census 2021, PCT, OpenStreetMap and weather), directing an AI assistant as you go and checking its output rather than trusting it.

It is more demanding than this session, and self-directed: expect to get stuck, and expect to work it out yourself.

What it adds:

  • Setup guides for Windows, macOS, locked-down laptops, GitHub Codespaces and Google Colab, plus a script that diagnoses what is wrong with your installation
  • Twelve self-checking programming drills and a “traceback safari” of six deliberately broken scripts that mark themselves
  • A full method for working with AI: how to specify a task, and the four checks to run before you believe the result
  • A worked Leeds atlas you can follow chapter by chapter, and a brief for building your own

Two things to know before you start:

  • Python only. There is no R version, so if you are on the R track, R for Data Science (2e) and Geocomputation with R are the better follow-ons.
  • Budget about 45 minutes for setup before week 1. The guides assume a local Python and VS Code, with Codespaces or Colab as the fallback if that is not possible.

The same text is published as a searchable website at https://chris-r-uol.github.io/python_learning/, but the repository is the copy to clone or download, since the site is built from it.

Advanced Topics to Explore

Once you’re comfortable with basics:

Data Science Skills

  • Version Control: Git and GitHub for collaboration
  • Reproducible Research: Quarto and reproducible notebooks for dissertations

Specialized Topics

  • Spatial Data Science: Working with spatial and transport data — see Geocomputation with Python and Geocomputation with R at geocompx.org
  • Machine Learning: Supervised and unsupervised learning
  • Deep Learning: Neural networks for complex patterns
  • Time Series Analysis: For temporal data
  • Network Analysis: For transport networks and routing

Final Tips

TipThe Best Way to Learn

Practice, practice, practice! Learning data science is like learning a language—you need to use it regularly to improve.

ImportantDon’t Get Overwhelmed

There’s a lot to learn, but you don’t need to learn everything at once. Pick one area, get comfortable, then expand gradually.

NoteStay Curious

The field of data science is constantly evolving. Stay curious, keep learning, and don’t be afraid to experiment!

Reuse