Datasets

To help you get started quickly, we offer several entry points into the dataset. Choose the path that best matches your application and available resources.

• Skip data pre-processing & start modeling

  • Best for rapid prototyping
  • Lightweight (runs in Colab)

  This option helps you understand the data structure and start building immediately, without needing to install or host a database.

  • Download generic sampled communities to build your own models
  • Run graph 101 (g101) Jupyter Notebook, or explore quickstart demo models

• Skip the ETL pipeline & sample your own communities

  • Best for specialized applications
  • Resource-intensive with long runtimes

  This option enables you to sample custom neighborhoods tailored to your specific application.

  • Import graph into a Neo4j database or restore the database dump
  • Documentation on sampling your own graphs

• Skip the Bitcoin node sync and chain traverse & use summary statistics

  • Best for macro analysis
  • Lightweight (Runs in Colab)

  This option provides longitudinal statistics of the Bitcoin blockchain, which you can study individually or use to augment multi-modal architectures.

  • Download block summary statistics
  • Documentation