ARCH — The estimation scenario uses a cleaned 9,020-row TAWOS dataset (SOLAR…
The estimation scenario uses a cleaned 9,020-row TAWOS dataset (SOLAR group) of agile backlog items parsed from a SQL dump; TAWOS story points are long-tailed (median 3.0), making MAE sensitive to sampled instances.
Source: puma docs/scenarios_reference.md; specs/datasets.spec.md · Traceability: corpus unit
PUMA-042· Confidence: repo-backed