prediction_baseline.py

import csv


INPUT_PATH = "data/tester.csv"
OUTPUT_PATH = "predictions.csv"


def to_float(row, key):
    return float(row[key])


def baseline_predict(row):
    pressure = to_float(row, "system_pressure")
    buffer_capacity = to_float(row, "buffer_capacity")
    intervention_lag = to_float(row, "intervention_lag")
    subsystem_coupling = to_float(row, "subsystem_coupling")
    drift_gradient = to_float(row, "drift_gradient")
    boundary_distance = to_float(row, "boundary_distance")
    intervention_leverage_score = to_float(row, "intervention_leverage_score")
    intervention_alignment_score = to_float(row, "intervention_alignment_score")
    side_effect_load = to_float(row, "side_effect_load")
    recovery_window_width = to_float(row, "recovery_window_width")
    local_relief_score = to_float(row, "local_relief_score")
    global_destabilization_risk = to_float(row, "global_destabilization_risk")

    rescue_signal = (
        0.30 * buffer_capacity
        + 0.25 * intervention_alignment_score
        + 0.15 * recovery_window_width
        + 0.10 * intervention_leverage_score
        + 0.10 * boundary_distance
        + 0.10 * (1.0 - intervention_lag)
    )

    collapse_signal = (
        0.25 * pressure
        + 0.20 * subsystem_coupling
        + 0.20 * drift_gradient
        + 0.15 * side_effect_load
        + 0.10 * global_destabilization_risk
        + 0.10 * (1.0 - boundary_distance)
    )

    penalty = 0.0

    if (
        local_relief_score >= 0.70
        and global_destabilization_risk >= 0.65
        and subsystem_coupling >= 0.65
    ):
        penalty += 0.15

    if (
        intervention_lag >= 0.70
        and recovery_window_width <= 0.30
        and boundary_distance <= 0.30
    ):
        penalty += 0.15

    score = rescue_signal - collapse_signal - penalty

    return 1 if score >= 0.00 else 0


def main():
    with open(INPUT_PATH, "r", newline="", encoding="utf-8") as f:
        reader = csv.DictReader(f)
        rows = list(reader)

    predictions = []
    for row in rows:
        predictions.append(
            {
                "scenario_id": row["scenario_id"],
                "prediction": baseline_predict(row),
            }
        )

    with open(OUTPUT_PATH, "w", newline="", encoding="utf-8") as f:
        writer = csv.DictWriter(f, fieldnames=["scenario_id", "prediction"])
        writer.writeheader()
        writer.writerows(predictions)

    print(f"Wrote {len(predictions)} predictions to {OUTPUT_PATH}")


if __name__ == "__main__":
    main()