Guides¶
Step-by-step walkthroughs for the most common tasks.
-
Run a session from disk
Load a raw session directory, run individual processors, and access the output DataFrames.
-
Query parquet files
Work with the export output using pandas, DuckDB, and Polars. Covers the session catalogue, flat cross-session tables, and per-session large tables.
-
Query from AWS S3
Point DuckDB or Polars directly at S3 — no download needed. Includes SSO credential setup and glob-based multi-session queries.
-
:material-stream:{ .lg .middle } Raw streams with contraqctor
Access the underlying behavioral streams directly via contraqctor. Useful for custom analysis beyond the built-in processors.