Additional Resources

Handy links, datasets, and tools to go deeper with DuckDB and analytics.

Additional Resources

Curated references to continue your DuckDB journey.

Official

• DuckDB Documentation: https://duckdb.org/docs/
• SQL Functions and Operators: https://duckdb.org/docs/sql/functions
• Python API: https://duckdb.org/docs/api/python
• Extensions: https://duckdb.org/docs/extensions/overview

Tutorials and datasets

• Seaborn sample datasets (small, friendly): https://github.com/mwaskom/seaborn-data
• FiveThirtyEight datasets (story-focused): https://github.com/fivethirtyeight/data
• UCI Machine Learning Repository (varied sizes): https://archive.ics.uci.edu/
• data.gov.uk, data.gov (public sector): curated open datasets by country
• NYC TLC trips (large): https://www.nyc.gov/site/tlc/about/tlc-trip-record-data.page
• Awesome DuckDB list: https://github.com/duckdb/awesome-duckdb

Starter query ideas

• Tidy a CSV into typed columns, export to Parquet
• Compute daily or monthly aggregates with DATE_TRUNC
• Compare categories with GROUP BY and ORDER BY
• Add a small dimension table (e.g., labels) and JOIN
• Try partitioned export and query one slice

Tools that pair well

• Parquet command line: parquet-tools
• Pandas or Polars for DataFrames
• VS Code + Jupyter for notebooks

Example repo structure

project/
  data/         # raw inputs (csv/parquet)
  exports/      # query outputs (parquet/png)
  analytics.duckdb
  queries/      # sql scripts
  notebooks/    # analysis notebooks
  README.md

Troubleshooting

• If remote reads fail, ensure httpfs is installed and loaded.
• For large scans, add predicates and select fewer columns.
• If memory errors occur, lower data size or set PRAGMA memory_limit.

Keep experimenting—small, composable queries add up to powerful analytics.

< Previous Next >

You can use the arrows  ← → on your keyboard to navigate between lessons.