Scientifiques des données Guidance for writing correct, performant retail analytics with the openretailscience Python package. Use when a task involves customer, basket, or transaction data and mentions any openretailscience analysis (RFM / HML / NLR / threshold segmentation, cross-shop Venn overlap, gain-loss switching, cohort retention, product association / market-basket, revenue-tree KPI decomposition, customer decision hierarchy, composite rank, haversine, customer-lifetime-value / Pareto-NBD / BTYD model-input prep), any of its plots (bar, line, area, scatter, histogram, waterfall, venn, heatmap, cohort, time, period-on-period, broken-timeline, price, index) or trendlines, its options/ColumnHelper configuration system, or connecting analyses to a database via Ibis. Also use whenever the user imports `openretailscience`.
2026-07-31