π Today's Summary β Ensemble v3.2 Predictions (CPU country)
All 0 active cohorts (cost β₯ $500). Ensemble v3.2 = age-aware blend (Cox-heavy <14d / 50-50 <60d / ENet-heavy β₯60d) + country CPU features (cpu, cpu_us, cost_us_share). Self-eval 0.9576. Retrained daily at 11:30 AM.
ποΈ All Cohorts β Click row to expand curve
Risk:
Age:
β cohorts
| Cohort | Age |
Cost | CPP |
ROAS D0 |
US% | Yr% |
Obs TD | Pred D360 | Pred BE |
Risk | Reason |
π¬ Last Week Summary
Pairs added in past 7 days. Empty = no cohorts reached a new milestone yet, or no archived predictions to compare. Backtest accumulates daily.
π Drift over time β MAPE per milestone
Rolling MAPE for each observable milestone. Lower = model more accurate. Trend up = drift, may trigger retrain.
π― Per-Milestone Backtest Pairs
β pairs
| Cohort | Milestone | Predict Date |
Predict Age | Pred |
Obs | Abs Err | MAPE |
π Retrain Log
| Date | Trigger |
OLD MAPE | NEW MAPE | Ξ |
Deployed | Reason |
Weekly retrain runs Sunday 13:00. Auto-rollback if NEW MAPE worse than OLD by >5pp.
π₯ Overall Leaderboard
| Rank | Agent | Total (weighted) | Total (simple) |
E1 Mature | E2 Mid | E3 Young |
E4 Cold | E5 OOT |
Weighted: Exam 3 Γ1.2, Exam 4-5 Γ1.5 (cold-start = real-world value).
Composite = 0.5Β·(1βMAPE_D360) + 0.25Β·BE_acc + 0.25Β·Risk_acc.
π― Strengths Heatmap
Green = strong, Red = weak. Identifies where each model excels.
π Per-Cohort Drill-Down
Prediction table (5 agents vs truth)
| Milestone | Truth |
π³ LGB | π ENet | β³ Cox |
π² Bayes | π¬ Gompertz |
π Methodology
- Data: 223 IIP555 cohorts (Nov 2025 β Jun 2026), stratified 171 train / 50 test.
- Features (51 total): ROAS curve (9), Cancel signal (8), Renewal (8), Refund (5), Geo US+EU (4), Package mix (5), Cost (5), Velocity (4), Meta (3).
- 5 Exams: cut-off day determines info visible β E1 D180 (easy) β E5 D1 (real-world).
- Scoring: composite = 0.5Β·(1βMAPE D360 clipped) + 0.25Β·(BE day within Β±30d) + 0.25Β·(Red/Yellow/Green risk class match).