Spaces:
Sleeping
Sleeping
File size: 19,848 Bytes
29ca14e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 | #!/usr/bin/env python3
"""
β οΈ DEVELOPER TOOL ONLY - NOT FOR END USERS
This script runs the prediction engine on completed races and measures accuracy
against actual results to verify if the 80% target is achievable.
FOR PREDICTIONS, USE THE STREAMLIT APP:
streamlit run app.py
Then access: http://localhost:8501
USAGE (Developers Only):
py scripts/verify_accuracy.py --season 2024
py scripts/verify_accuracy.py --all
OUTPUTS:
- Console summary of accuracy metrics
- Detailed JSON report: accuracy_report_YYYYMMDD.json
"""
import json
import logging
import sys
from collections import defaultdict
from datetime import datetime
from pathlib import Path
from typing import Dict, List, Any, Optional
# Add project root to Python path
project_root = Path(__file__).parent.parent
sys.path.insert(0, str(project_root))
import numpy as np
# Setup logging
logging.basicConfig(
level=logging.INFO,
format="%(asctime)s [%(levelname)s] %(message)s",
handlers=[logging.StreamHandler(sys.stdout)]
)
logger = logging.getLogger(__name__)
def load_historical_data(season: int) -> List[Dict[str, Any]]:
"""Load actual race results for a given season."""
try:
if season == 2026:
from src.data.season_2026 import SEASON_RESULTS_2026
logger.info(f"Loaded {len(SEASON_RESULTS_2026)} races from 2026 season data")
return SEASON_RESULTS_2026
elif season in [2024, 2025]:
# Try FastF1 for historical seasons
try:
from src.data.fastf1_integration import load_entire_season
races = load_entire_season(season)
logger.info(f"Loaded {len(races)} races from {season} via FastF1")
return races
except Exception as e:
logger.warning(f"FastF1 failed for {season}: {e}")
logger.info("Using fallback: No historical data available for this season")
return []
else:
logger.warning(f"No data source configured for season {season}")
return []
except ImportError as e:
logger.error(f"Failed to import season data: {e}")
return []
def run_prediction(circuit_id: str, rain_prob: Optional[float] = None,
n_simulations: int = 5000, use_qualifying_grid: bool = False) -> Dict:
"""Run prediction for a specific circuit."""
from src.engine.predictor import predict as run_predict, PredictionRequest
try:
result = run_predict(
PredictionRequest(
circuit_id=circuit_id,
rain_probability=rain_prob,
n_simulations=n_simulations,
seed=None, # No fixed seed for realistic variance
grid_overrides={}, # Empty for pre-qualifying predictions
use_live_data=use_qualifying_grid, # Use live data if qualifying available
)
)
return result
except Exception as e:
logger.error(f"Prediction failed for {circuit_id}: {e}")
return {"predictions": [], "error": str(e)}
def calculate_accuracy_metrics(predictions: List[Dict], actual_results: List[Dict]) -> Dict[str, Any]:
"""
Calculate comprehensive accuracy metrics comparing predictions to actual results.
Metrics:
- Top-3 accuracy: How many of top-3 predicted drivers finished in top-3
- Winner prediction: Was the actual winner in predicted top-3?
- Points finisher accuracy: How many of top-10 predicted actually scored points
- Position correlation: Spearman-like correlation between predicted and actual positions
- Mean absolute error: Average position prediction error
"""
if not predictions or not actual_results:
return {"error": "Missing predictions or actual results"}
# Build lookup maps
pred_by_driver = {p["driver_id"]: p for p in predictions}
actual_by_driver = {r["driver"]: r for r in actual_results}
# Sort by predicted and actual positions
sorted_preds = sorted(predictions, key=lambda x: x.get("expected_position_float", x.get("predicted_position", 999)))
sorted_actuals = sorted(actual_results, key=lambda x: x.get("position", 999))
# Extract driver IDs in order
pred_top3 = [p["driver_id"] for p in sorted_preds[:3]]
actual_top3 = [r["driver"] for r in sorted_actuals[:3]]
pred_top10 = [p["driver_id"] for p in sorted_preds[:10]]
actual_top10 = [r["driver"] for r in sorted_actuals[:10]]
# Calculate metrics
metrics = {
# Top-3 accuracy
"top3_intersection": list(set(pred_top3) & set(actual_top3)),
"top3_correct_count": len(set(pred_top3) & set(actual_top3)),
"top3_accuracy_pct": len(set(pred_top3) & set(actual_top3)) / 3 * 100,
# Winner prediction
"actual_winner": sorted_actuals[0]["driver"] if sorted_actuals else None,
"winner_in_pred_top3": sorted_actuals[0]["driver"] in pred_top3 if sorted_actuals else False,
"winner_predicted_position": pred_by_driver.get(sorted_actuals[0]["driver"], {}).get("expected_position_float", 999) if sorted_actuals else None,
# Points finisher accuracy (top 10)
"points_intersection": list(set(pred_top10) & set(actual_top10)),
"points_correct_count": len(set(pred_top10) & set(actual_top10)),
"points_accuracy_pct": len(set(pred_top10) & set(actual_top10)) / min(10, len(actual_top10)) * 100 if actual_top10 else 0,
# Position errors
"position_errors": {},
"absolute_position_errors": [],
}
# Calculate position-by-position errors
for driver_id in pred_by_driver:
if driver_id in actual_by_driver:
pred_pos = pred_by_driver[driver_id].get("expected_position_float", 999)
actual_pos = actual_by_driver[driver_id].get("position", 999)
error = abs(pred_pos - actual_pos)
metrics["position_errors"][driver_id] = {
"predicted": round(pred_pos, 2),
"actual": actual_pos,
"error": round(error, 2)
}
metrics["absolute_position_errors"].append(error)
# Aggregate position metrics
if metrics["absolute_position_errors"]:
metrics["mean_absolute_error"] = sum(metrics["absolute_position_errors"]) / len(metrics["absolute_position_errors"])
metrics["median_absolute_error"] = np.median(metrics["absolute_position_errors"])
metrics["max_absolute_error"] = max(metrics["absolute_position_errors"])
# Percentage within certain thresholds
within_1 = sum(1 for e in metrics["absolute_position_errors"] if e <= 1) / len(metrics["absolute_position_errors"]) * 100
within_2 = sum(1 for e in metrics["absolute_position_errors"] if e <= 2) / len(metrics["absolute_position_errors"]) * 100
within_3 = sum(1 for e in metrics["absolute_position_errors"] if e <= 3) / len(metrics["absolute_position_errors"]) * 100
metrics["pct_within_1_position"] = within_1
metrics["pct_within_2_positions"] = within_2
metrics["pct_within_3_positions"] = within_3
return metrics
def generate_summary_report(all_race_metrics: List[Dict]) -> Dict[str, Any]:
"""Generate aggregate summary across all races."""
if not all_race_metrics:
return {"error": "No race metrics to summarize"}
n_races = len(all_race_metrics)
# Aggregate top-3 accuracy
avg_top3_accuracy = sum(m["top3_accuracy_pct"] for m in all_race_metrics) / n_races
top3_accuracies = [m["top3_accuracy_pct"] for m in all_race_metrics]
# Aggregate winner prediction
winner_in_top3_count = sum(1 for m in all_race_metrics if m.get("winner_in_pred_top3", False))
winner_accuracy = winner_in_top3_count / n_races * 100
# Aggregate points accuracy
avg_points_accuracy = sum(m["points_accuracy_pct"] for m in all_race_metrics) / n_races
# Aggregate position errors
all_mean_errors = [m.get("mean_absolute_error", 0) for m in all_race_metrics if "mean_absolute_error" in m]
overall_mean_error = sum(all_mean_errors) / len(all_mean_errors) if all_mean_errors else None
all_median_errors = [m.get("median_absolute_error", 0) for m in all_race_metrics if "median_absolute_error" in m]
overall_median_error = np.median(all_median_errors) if all_median_errors else None
# Position threshold accuracies
avg_within_1 = sum(m.get("pct_within_1_position", 0) for m in all_race_metrics) / n_races
avg_within_2 = sum(m.get("pct_within_2_positions", 0) for m in all_race_metrics) / n_races
avg_within_3 = sum(m.get("pct_within_3_positions", 0) for m in all_race_metrics) / n_races
summary = {
"total_races_analyzed": n_races,
"analysis_date": datetime.now().isoformat(),
"top3_accuracy": {
"average_pct": round(avg_top3_accuracy, 2),
"min_pct": round(min(top3_accuracies), 2),
"max_pct": round(max(top3_accuracies), 2),
"std_dev": round(np.std(top3_accuracies), 2),
"target_80_pct": "β
ACHIEVED" if avg_top3_accuracy >= 80 else f"β GAP: {80 - avg_top3_accuracy:.1f}%",
},
"winner_prediction": {
"accuracy_pct": round(winner_accuracy, 2),
"correct_count": winner_in_top3_count,
"total_races": n_races,
},
"points_finisher_accuracy": {
"average_pct": round(avg_points_accuracy, 2),
"target_70_pct": "β
ACHIEVED" if avg_points_accuracy >= 70 else f"β GAP: {70 - avg_points_accuracy:.1f}%",
},
"position_prediction": {
"mean_absolute_error": round(overall_mean_error, 2) if overall_mean_error else None,
"median_absolute_error": round(overall_median_error, 2) if overall_median_error else None,
"pct_within_1_position": round(avg_within_1, 2),
"pct_within_2_positions": round(avg_within_2, 2),
"pct_within_3_positions": round(avg_within_3, 2),
},
"verdict": {
"meets_80_pct_target": avg_top3_accuracy >= 80,
"recommendation": _generate_recommendation(avg_top3_accuracy, winner_accuracy, avg_points_accuracy),
}
}
return summary
def _generate_recommendation(top3_acc: float, winner_acc: float, points_acc: float) -> str:
"""Generate actionable recommendation based on accuracy metrics."""
if top3_acc >= 80:
return "π EXCELLENT! Model meets 80% accuracy target. Continue monitoring and consider deploying to production."
elif top3_acc >= 70:
improvements = []
if winner_acc < 75:
improvements.append("- Improve winner prediction (currently {:.1f}%)".format(winner_acc))
if points_acc < 70:
improvements.append("- Improve points finisher prediction (currently {:.1f}%)".format(points_acc))
return f"β οΈ GOOD but needs refinement. Top-3 accuracy is {top3_acc:.1f}%.\nRecommendations:\n" + "\n".join(improvements) + "\n- Consider using real qualifying grid data (+15-20% improvement expected)"
else:
return f"β BELOW TARGET. Top-3 accuracy is {top3_acc:.1f}%.\nCritical actions needed:\n- Ensure real qualifying grid is used for Sunday predictions\n- Train isotonic calibration on historical data\n- Review feature weights in feature_engineering.py\n- Increase simulation count to 10,000+"
def verify_accuracy(seasons: List[int], output_file: str = None, use_qualifying_data: bool = True):
"""Run full accuracy verification across specified seasons."""
logger.info("=" * 80)
logger.info("F1 PREDICTOR ACCURACY VERIFICATION")
logger.info("=" * 80)
logger.info(f"Seasons to test: {seasons}")
logger.info(f"Use qualifying grid data: {'Yes' if use_qualifying_data else 'No'}")
logger.info("=" * 80 + "\n")
all_race_metrics = []
for season in seasons:
logger.info(f"\n{'='*80}")
logger.info(f"Testing Season {season}")
logger.info(f"{'='*80}\n")
races = load_historical_data(season)
if not races:
logger.warning(f"No races found for season {season}. Skipping.\n")
continue
for i, race in enumerate(races, 1):
circuit_id = race.get("circuit", race.get("location", "unknown"))
race_name = race.get("name", circuit_id)
round_num = race.get("round", "?")
logger.info(f"[{i}/{len(races)}] Round {round_num}: {race_name} ({circuit_id})")
try:
# Run prediction
result = run_prediction(
circuit_id=circuit_id,
rain_prob=race.get("rain_probability_typical", 0.2),
n_simulations=5000,
use_qualifying_grid=use_qualifying_data,
)
if "error" in result:
logger.warning(f" β οΈ Prediction error: {result['error']}")
continue
predictions = result.get("predictions", [])
actual_results = race.get("results", [])
if not predictions or not actual_results:
logger.warning(f" β οΈ Missing data. Skipping.")
continue
# Calculate accuracy
metrics = calculate_accuracy_metrics(predictions, actual_results)
# Add race metadata
metrics["season"] = season
metrics["round"] = round_num
metrics["circuit"] = circuit_id
metrics["race_name"] = race_name
all_race_metrics.append(metrics)
# Log per-race results
logger.info(f" β Top-3 Accuracy: {metrics['top3_accuracy_pct']:.1f}%")
logger.info(f" β Winner in Pred Top-3: {'Yes' if metrics['winner_in_pred_top3'] else 'No'}")
logger.info(f" β Points Finishers: {metrics['points_accuracy_pct']:.1f}%")
if "mean_absolute_error" in metrics:
logger.info(f" β Mean Position Error: {metrics['mean_absolute_error']:.2f}")
except Exception as e:
logger.error(f" β Failed: {e}")
import traceback
logger.debug(traceback.format_exc())
# Generate summary
logger.info(f"\n{'='*80}")
logger.info("ACCURACY SUMMARY")
logger.info(f"{'='*80}\n")
summary = generate_summary_report(all_race_metrics)
# Print summary
print("\n" + "=" * 80)
print("FINAL ACCURACY REPORT")
print("=" * 80)
if not summary or "total_races_analyzed" not in summary:
print("\nβ οΈ No races were analyzed. Possible reasons:")
print(" β’ No historical data available for selected seasons")
print(" β’ FastF1 connection failed")
print(" β’ All races were skipped due to errors")
print("\nπ‘ Recommendations:")
print(" 1. Check your internet connection for FastF1")
print(" 2. Try with --season 2026 only (uses local data)")
print(" 3. Verify FastF1 is installed: py -m pip show fastf1")
print("=" * 80 + "\n")
logger.warning("No accuracy data generated. Check data sources.")
return report
print(f"\nTotal Races Analyzed: {summary['total_races_analyzed']}")
print(f"\nπ TOP-3 ACCURACY:")
print(f" Average: {summary['top3_accuracy']['average_pct']:.1f}%")
print(f" Range: {summary['top3_accuracy']['min_pct']:.1f}% - {summary['top3_accuracy']['max_pct']:.1f}%")
print(f" Target (80%): {summary['top3_accuracy']['target_80_pct']}")
print(f"\nπ WINNER PREDICTION:")
print(f" Accuracy: {summary['winner_prediction']['accuracy_pct']:.1f}%")
print(f" Correct: {summary['winner_prediction']['correct_count']}/{summary['winner_prediction']['total_races']}")
print(f"\nπ― POINTS FINISHER ACCURACY:")
print(f" Average: {summary['points_finisher_accuracy']['average_pct']:.1f}%")
print(f" Target (70%): {summary['points_finisher_accuracy']['target_70_pct']}")
print(f"\nπ POSITION PREDICTION:")
if summary['position_prediction']['mean_absolute_error']:
print(f" Mean Absolute Error: {summary['position_prediction']['mean_absolute_error']:.2f} positions")
print(f" Median Absolute Error: {summary['position_prediction']['median_absolute_error']:.2f} positions")
print(f" Within 1 Position: {summary['position_prediction']['pct_within_1_position']:.1f}%")
print(f" Within 2 Positions: {summary['position_prediction']['pct_within_2_positions']:.1f}%")
print(f" Within 3 Positions: {summary['position_prediction']['pct_within_3_positions']:.1f}%")
print(f"\nπ‘ RECOMMENDATION:")
print(summary['verdict']['recommendation'])
print("=" * 80 + "\n")
# Save detailed report
report = {
"summary": summary,
"per_race_metrics": all_race_metrics,
}
if output_file:
with open(output_file, "w") as f:
json.dump(report, f, indent=2, default=str)
logger.info(f"Detailed report saved to: {output_file}")
return report
def main():
import argparse
parser = argparse.ArgumentParser(description="Verify F1 Predictor accuracy against historical data")
parser.add_argument("--season", type=int, nargs="+", help="Season(s) to test (e.g., --season 2024 2025)")
parser.add_argument("--all", action="store_true", help="Test all available seasons (2024, 2025, 2026)")
parser.add_argument("--output", type=str, default=None, help="Output JSON report file path")
parser.add_argument("--no-qualifying", action="store_true", help="Disable use of qualifying grid data (baseline test)")
args = parser.parse_args()
# Determine seasons to test
if args.all:
seasons = [2024, 2025, 2026]
elif args.season:
seasons = args.season
else:
seasons = [2024, 2025] # Default: historical seasons
# Determine output file
if not args.output:
timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
args.output = f"accuracy_report_{timestamp}.json"
# Run verification
use_qualifying = not args.no_qualifying
start_time = datetime.now()
try:
report = verify_accuracy(seasons, output_file=args.output, use_qualifying_data=use_qualifying)
elapsed = (datetime.now() - start_time).total_seconds()
logger.info(f"\nVerification completed in {elapsed:.1f} seconds")
# Exit code based on accuracy
summary = report.get("summary", {})
top3_acc = summary.get("top3_accuracy", {}).get("average_pct", 0)
if top3_acc >= 80:
logger.info("β
SUCCESS: 80% accuracy target achieved!")
sys.exit(0)
elif top3_acc >= 70:
logger.warning("β οΈ WARNING: Accuracy below 80% target but above 70% minimum")
sys.exit(0)
else:
logger.error("β FAILURE: Accuracy below 70%. Model needs significant improvement.")
sys.exit(1)
except Exception as e:
logger.error(f"Verification failed: {e}")
import traceback
traceback.print_exc()
sys.exit(1)
if __name__ == "__main__":
main()
|