← All runsCompleted
demand-forecast-lgbm-v3
Weekly demand forecast · LightGBM (lags + weather + holidays)
Metrics vs baselines
| Metric | Model | Naive baseline | Current method | Δ vs best baseline |
|---|---|---|---|---|
| WAPE | 11.8% | 19.4% | 16.2% | −4.4 pts |
| MAE (units) | 8.3 | 13.1 | 11.0 | −2.7 |
| Stockout events (held-out) | 41 | 68 | 59 | −18 |
Baseline gate
To pass evaluation, the model must beat both the naive baseline and your current method.
Gate passed
Feature importance (SHAP)
lag_1 (last week sales)31.0%
lag_4 (same week, last month)22.0%
temp_c (avg)14.0%
holiday (QC)11.0%
promo_active9.0%
price_index7.0%
site_id6.0%
Model card
- Task
- Forecasting
- Target
- units_sold_next_7d per site × SKU
- Validation
- Time-based split — last 12 weeks held out
- Data version
- borealdump-2026-09-17
- Model
- LightGBM (lags + weather + holidays)
- Training cost
- $4.20 CAD