Spaces:
Running
Running
File size: 19,082 Bytes
c095e08 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 |
"""
NBA ML Prediction System - Prediction Tracker
==============================================
Tracks predictions and measures accuracy using ChromaDB Cloud with local fallback.
"""
import logging
from datetime import datetime
from typing import Dict, List, Optional
import json
import hashlib
from pathlib import Path
from src.config import CHROMADB_CONFIG, PROCESSED_DATA_DIR
logger = logging.getLogger(__name__)
class PredictionTracker:
"""
Tracks predictions and measures accuracy over time.
Uses ChromaDB Cloud if available, otherwise falls back to local JSON storage.
Stores predictions before games and updates with results after completion.
Provides accuracy statistics by team, confidence level, and over time.
"""
def __init__(self):
"""Initialize storage - try ChromaDB Cloud, fallback to local JSON."""
self.collection = None
self.client = None
self._use_local = False
self._local_file = PROCESSED_DATA_DIR / "predictions_local.json"
self._local_data: List[Dict] = []
# Try ChromaDB Cloud first
try:
import chromadb
# Try CloudClient (the official way for Chroma Cloud)
self.client = chromadb.CloudClient(
tenant=CHROMADB_CONFIG.tenant,
database=CHROMADB_CONFIG.database,
api_key=CHROMADB_CONFIG.api_key,
)
# Get or create collection for predictions
self.collection = self.client.get_or_create_collection(
name=CHROMADB_CONFIG.collection_name,
metadata={"description": "NBA game predictions with results"}
)
logger.info("Connected to ChromaDB Cloud successfully")
except Exception as e:
logger.warning(f"ChromaDB Cloud unavailable ({e}), using local JSON storage")
self._use_local = True
self._load_local_data()
def _generate_id(self, game_id: str, prediction_date: str) -> str:
"""Generate unique ID for a prediction."""
return hashlib.md5(f"{game_id}_{prediction_date}".encode()).hexdigest()
def _load_local_data(self):
"""Load predictions from local JSON file."""
if self._local_file.exists():
try:
with open(self._local_file, 'r') as f:
self._local_data = json.load(f)
except Exception as e:
logger.error(f"Failed to load local data: {e}")
self._local_data = []
else:
self._local_data = []
def _save_local_data(self):
"""Save predictions to local JSON file."""
try:
with open(self._local_file, 'w') as f:
json.dump(self._local_data, f, indent=2)
except Exception as e:
logger.error(f"Failed to save local data: {e}")
def _find_local_prediction(self, game_id: str) -> Optional[int]:
"""Find index of prediction by game_id in local data."""
for i, pred in enumerate(self._local_data):
if pred.get("game_id") == game_id:
return i
return None
def save_prediction(self, game_id: str, prediction: Dict) -> bool:
"""
Store a prediction before game starts.
Args:
game_id: NBA game ID
prediction: Dict with home_team, away_team, predicted_winner,
home_win_prob, confidence, etc.
Returns:
True if saved successfully
"""
now = datetime.now().isoformat()
doc_id = self._generate_id(game_id, now[:10])
# Prepare metadata
metadata = {
"id": doc_id,
"game_id": game_id,
"game_date": prediction.get("game_date", now[:10]),
"home_team": prediction.get("home_team", ""),
"away_team": prediction.get("away_team", ""),
"predicted_winner": prediction.get("predicted_winner", ""),
"home_win_prob": float(prediction.get("home_win_probability", 0.5)),
"away_win_prob": float(prediction.get("away_win_probability", 0.5)),
"confidence": prediction.get("confidence", "medium"),
"home_elo": float(prediction.get("home_elo", 1500)),
"away_elo": float(prediction.get("away_elo", 1500)),
"actual_winner": "", # Empty until game completes
"is_correct": -1, # -1 = pending, 0 = wrong, 1 = correct
"created_at": now,
"updated_at": now,
}
# Use local storage if ChromaDB not available
if self._use_local:
try:
# Check if exists and update, otherwise append
idx = self._find_local_prediction(game_id)
if idx is not None:
self._local_data[idx] = metadata
else:
self._local_data.append(metadata)
self._save_local_data()
logger.info(f"Saved prediction for game {game_id} (local)")
return True
except Exception as e:
logger.error(f"Failed to save prediction locally: {e}")
return False
# Use ChromaDB Cloud
if not self.collection:
logger.warning("ChromaDB not available, prediction not saved")
return False
try:
# Document text for semantic search
doc_text = (
f"NBA Game: {prediction.get('away_team')} @ {prediction.get('home_team')} "
f"on {metadata['game_date']}. "
f"Predicted winner: {metadata['predicted_winner']} "
f"with {metadata['confidence']} confidence "
f"({metadata['home_win_prob']:.1%} home win probability)"
)
# Upsert (update if exists, insert if not)
self.collection.upsert(
ids=[doc_id],
documents=[doc_text],
metadatas=[metadata]
)
logger.info(f"Saved prediction for game {game_id}")
return True
except Exception as e:
logger.error(f"Failed to save prediction: {e}")
return False
def update_result(self, game_id: str, actual_winner: str,
home_score: int = 0, away_score: int = 0) -> bool:
"""
Update prediction with actual game result.
Args:
game_id: NBA game ID
actual_winner: Team tricode of actual winner
home_score: Final home score
away_score: Final away score
Returns:
True if updated successfully
"""
# Handle local storage
if self._use_local:
try:
idx = self._find_local_prediction(game_id)
if idx is None:
logger.warning(f"No prediction found for game {game_id}")
return False
pred = self._local_data[idx]
predicted_winner = pred.get("predicted_winner", "")
is_correct = 1 if predicted_winner == actual_winner else 0
pred["actual_winner"] = actual_winner
pred["is_correct"] = is_correct
pred["home_score"] = home_score
pred["away_score"] = away_score
pred["updated_at"] = datetime.now().isoformat()
self._local_data[idx] = pred
self._save_local_data()
result_text = "CORRECT ✓" if is_correct else "WRONG ✗"
logger.info(f"Updated result for game {game_id}: {result_text} (local)")
return True
except Exception as e:
logger.error(f"Failed to update result locally: {e}")
return False
# Handle ChromaDB Cloud
if not self.collection:
return False
try:
# Find the prediction for this game
results = self.collection.get(
where={"game_id": game_id},
include=["metadatas", "documents"]
)
if not results["ids"]:
logger.warning(f"No prediction found for game {game_id}")
return False
doc_id = results["ids"][0]
metadata = results["metadatas"][0]
# Check if prediction was correct
predicted_winner = metadata.get("predicted_winner", "")
is_correct = 1 if predicted_winner == actual_winner else 0
# Update metadata
metadata["actual_winner"] = actual_winner
metadata["is_correct"] = is_correct
metadata["home_score"] = home_score
metadata["away_score"] = away_score
metadata["updated_at"] = datetime.now().isoformat()
# Update document text
result_text = "CORRECT ✓" if is_correct else "WRONG ✗"
doc_text = (
f"NBA Game: {metadata['away_team']} @ {metadata['home_team']}. "
f"Predicted: {predicted_winner}, Actual: {actual_winner}. "
f"Result: {result_text}"
)
self.collection.update(
ids=[doc_id],
documents=[doc_text],
metadatas=[metadata]
)
logger.info(f"Updated result for game {game_id}: {result_text}")
return True
except Exception as e:
logger.error(f"Failed to update result: {e}")
return False
def get_prediction(self, game_id: str) -> Optional[Dict]:
"""Get prediction for a specific game."""
# Handle local storage
if self._use_local:
idx = self._find_local_prediction(game_id)
if idx is not None:
return self._local_data[idx]
return None
# Handle ChromaDB
if not self.collection:
return None
try:
results = self.collection.get(
where={"game_id": game_id},
include=["metadatas"]
)
if results["ids"]:
return results["metadatas"][0]
return None
except Exception as e:
logger.error(f"Failed to get prediction: {e}")
return None
def get_recent_predictions(self, n: int = 20) -> List[Dict]:
"""Get N most recent predictions with results."""
# Handle local storage
if self._use_local:
predictions = sorted(
self._local_data,
key=lambda x: x.get("created_at", ""),
reverse=True
)
return predictions[:n]
# Handle ChromaDB
if not self.collection:
return []
try:
# Get all predictions and sort by date
results = self.collection.get(
include=["metadatas"]
)
if not results["ids"]:
return []
predictions = results["metadatas"]
# Sort by created_at descending
predictions.sort(key=lambda x: x.get("created_at", ""), reverse=True)
return predictions[:n]
except Exception as e:
logger.error(f"Failed to get recent predictions: {e}")
return []
def _calculate_accuracy_from_predictions(self, predictions: List[Dict]) -> Dict:
"""Calculate accuracy stats from a list of predictions."""
if not predictions:
return {
"total_predictions": 0,
"completed_games": 0,
"correct_predictions": 0,
"overall_accuracy": 0.0,
"by_confidence": {},
"by_team": {},
}
# Filter to completed games only
completed = [p for p in predictions if p.get("is_correct", -1) >= 0]
correct = [p for p in completed if p.get("is_correct") == 1]
# By confidence level
confidence_stats = {}
for conf in ["high", "medium", "low"]:
conf_preds = [p for p in completed if p.get("confidence") == conf]
conf_correct = [p for p in conf_preds if p.get("is_correct") == 1]
if conf_preds:
confidence_stats[conf] = {
"total": len(conf_preds),
"correct": len(conf_correct),
"accuracy": len(conf_correct) / len(conf_preds)
}
# By team predicted
team_stats = {}
for pred in completed:
team = pred.get("predicted_winner", "")
if team not in team_stats:
team_stats[team] = {"total": 0, "correct": 0}
team_stats[team]["total"] += 1
if pred.get("is_correct") == 1:
team_stats[team]["correct"] += 1
for team in team_stats:
total = team_stats[team]["total"]
team_stats[team]["accuracy"] = team_stats[team]["correct"] / total if total > 0 else 0
return {
"total_predictions": len(predictions),
"completed_games": len(completed),
"correct_predictions": len(correct),
"overall_accuracy": len(correct) / len(completed) if completed else 0.0,
"by_confidence": confidence_stats,
"by_team": team_stats,
}
def get_accuracy_stats(self) -> Dict:
"""
Calculate comprehensive accuracy statistics.
Returns:
Dict with overall accuracy, by confidence, by team
"""
# Handle local storage
if self._use_local:
return self._calculate_accuracy_from_predictions(self._local_data)
# Handle ChromaDB
if not self.collection:
return {
"total_predictions": 0,
"completed_games": 0,
"correct_predictions": 0,
"overall_accuracy": 0.0,
"by_confidence": {},
"by_team": {},
}
try:
results = self.collection.get(include=["metadatas"])
if not results["ids"]:
return {
"total_predictions": 0,
"completed_games": 0,
"correct_predictions": 0,
"overall_accuracy": 0.0,
"by_confidence": {},
"by_team": {},
}
predictions = results["metadatas"]
# Filter to completed games only
completed = [p for p in predictions if p.get("is_correct", -1) >= 0]
correct = [p for p in completed if p.get("is_correct") == 1]
# By confidence level
confidence_stats = {}
for conf in ["high", "medium", "low"]:
conf_preds = [p for p in completed if p.get("confidence") == conf]
conf_correct = [p for p in conf_preds if p.get("is_correct") == 1]
if conf_preds:
confidence_stats[conf] = {
"total": len(conf_preds),
"correct": len(conf_correct),
"accuracy": len(conf_correct) / len(conf_preds)
}
# By team predicted
team_stats = {}
for pred in completed:
team = pred.get("predicted_winner", "")
if team not in team_stats:
team_stats[team] = {"total": 0, "correct": 0}
team_stats[team]["total"] += 1
if pred.get("is_correct") == 1:
team_stats[team]["correct"] += 1
for team in team_stats:
total = team_stats[team]["total"]
team_stats[team]["accuracy"] = team_stats[team]["correct"] / total if total > 0 else 0
return {
"total_predictions": len(predictions),
"completed_games": len(completed),
"correct_predictions": len(correct),
"overall_accuracy": len(correct) / len(completed) if completed else 0.0,
"by_confidence": confidence_stats,
"by_team": team_stats,
}
except Exception as e:
logger.error(f"Failed to get accuracy stats: {e}")
return {
"total_predictions": 0,
"completed_games": 0,
"correct_predictions": 0,
"overall_accuracy": 0.0,
"by_confidence": {},
"by_team": {},
"error": str(e)
}
def get_pending_predictions(self) -> List[Dict]:
"""Get predictions for games not yet completed."""
if not self.collection:
return []
try:
results = self.collection.get(
where={"is_correct": -1},
include=["metadatas"]
)
return results.get("metadatas", [])
except Exception as e:
logger.error(f"Failed to get pending predictions: {e}")
return []
# =============================================================================
# CLI INTERFACE
# =============================================================================
if __name__ == "__main__":
logging.basicConfig(level=logging.INFO)
tracker = PredictionTracker()
print("\n=== Prediction Tracker Stats ===\n")
stats = tracker.get_accuracy_stats()
print(f"Total Predictions: {stats['total_predictions']}")
print(f"Completed Games: {stats['completed_games']}")
print(f"Correct Predictions: {stats['correct_predictions']}")
print(f"Overall Accuracy: {stats['overall_accuracy']:.1%}")
if stats['by_confidence']:
print("\nBy Confidence Level:")
for conf, data in stats['by_confidence'].items():
print(f" {conf.upper()}: {data['correct']}/{data['total']} ({data['accuracy']:.1%})")
print("\n=== Recent Predictions ===\n")
recent = tracker.get_recent_predictions(5)
for pred in recent:
status = "✓" if pred.get("is_correct") == 1 else "✗" if pred.get("is_correct") == 0 else "⏳"
print(f"{status} {pred.get('away_team')} @ {pred.get('home_team')} - Predicted: {pred.get('predicted_winner')}")
|