table = political_risk_score(-5.0, [-1.0, -0.5, 0.0, 0.5, 1.0, 0.5], 3.0, 0.0, 18.0) stressed = political_risk_score(-10.0, [-2.0, -2.0, -2.0, -2.0, -2.0, -2.0], 0.0 ...
Retraining policy for forward test: one-shot only (no retraining inside the 20% window). Freeze model, hyperparameters, and thresholds before touching the 20% block. Run a single final evaluation on ...