Data Users

Data users want to access and analyze published dsgrid datasets. You do not need to install dsgrid or work with the registry directly — the published data is available as Parquet files that can be read with standard tools like Python, R, or Tableau.

What’s Available

dsgrid publishes assembled, quality-checked energy demand datasets. These datasets combine contributions from many modeling teams into a unified dimensional structure so that you can query across sectors, geographies, time periods, and scenarios in a single analysis.

See Published Datasets for the current catalog.

Getting Started

  1. Browse the catalog — Review Published Datasets to find the dataset you need.

  2. Follow the tutorial — The Query a Dataset with Python tutorial walks through loading, filtering, and exporting data using Python and DuckDB.

  3. Visualize results — See Visualize Data with Tableau for creating dashboards.

Requirements

To work with published data you will need:

  • Python 3.11+ with duckdb and pyarrow (for the Python tutorial), or

  • Tableau with a Spark JDBC driver (for Tableau visualization), or

  • Any tool that can read Parquet files

Access to NLR HPC is required for some datasets that are not yet available on OEDI. Check the dataset catalog for access details.

Next Steps