featurizer × chicago311

Which 311 requests will resolve slowly? (requests ← area/type demand streams)

1 · Full-cohort materialization (v1.0.0)

Each variant materialized through to_dataframe(connection=…) at one as-of date (2020-01-01 — the day after the last knowledge date, so every event is under the window); no permanent writes. Variant definitions and the recovered wide transformer set: benchmarks/final_matrix.py.

variantfeaturesshardsrows dup-namessecondswas (v0.8.0)status
narrow28130,65400.80.8materialized
all-agg191130,65405.76.0materialized
wide907130,654047.549.2materialized