Spaces:
Running
Running
File size: 12,128 Bytes
e7f1d57 |
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 |
# Dependencies
import time
from typing import List
from typing import Dict
from typing import Tuple
from pathlib import Path
from typing import Callable
from utils.logger import get_logger
from config.settings import settings
from config.schemas import AnalysisResult
from concurrent.futures import TimeoutError
from concurrent.futures import as_completed
from config.constants import DetectionStatus
from config.schemas import BatchAnalysisResult
from metrics.aggregator import MetricsAggregator
from concurrent.futures import ThreadPoolExecutor
from features.threshold_manager import ThresholdManager
# Setup Logging
logger = get_logger(__name__)
class BatchProcessor:
"""
Process multiple images in parallel or sequential mode
Features:
---------
- Parallel processing using ThreadPoolExecutor
- Sequential fallback for single images or disabled parallel mode
- Automatic error handling and recovery
- Progress tracking and logging
"""
def __init__(self, threshold_manager: ThresholdManager):
"""
Initialize Batch Processor
"""
# Instantiate threshold manager
self.threshold_manager = threshold_manager
# Initialize aggregator
self.aggregator = MetricsAggregator(threshold_manager = threshold_manager)
# Fix number of workers
self.max_workers = settings.MAX_WORKERS if settings.PARALLEL_PROCESSING else 1
logger.info(f"BatchProcessor initialized with max_workers={self.max_workers}, parallel={settings.PARALLEL_PROCESSING}")
def process_batch(self, image_files: List[Dict[str, any]], on_progress: Callable[[int, int, str], None] | None = None) -> BatchAnalysisResult:
"""
Process multiple images with automatic parallel/sequential switching
Arguments:
----------
image_files { list } : List of dicts with keys:
- 'path' : Path object
- 'filename' : str
- 'size' : tuple (width, height)
on_progress { Callablel } : Optional callback invoked after each image is processed
Returns:
--------
{ BatchAnalysisResult } : Complete batch analysis result
"""
start_time = time.time()
total_images = len(image_files)
logger.info(f"Starting batch processing of {total_images} images")
# Validate input
if (total_images == 0):
logger.warning("Empty batch provided")
return self._create_empty_batch_result()
if (total_images > settings.MAX_BATCH_SIZE):
logger.error(f"Batch size {total_images} exceeds maximum {settings.MAX_BATCH_SIZE}")
raise ValueError(f"Batch size {total_images} exceeds maximum allowed {settings.MAX_BATCH_SIZE}")
# Choose processing strategy
if (settings.PARALLEL_PROCESSING and (total_images > 1)):
results, failed = self._process_parallel(image_files = image_files,
on_progress = on_progress,
)
else:
results, failed = self._process_sequential(image_files = image_files,
on_progress = on_progress,
)
total_time = time.time() - start_time
# Create batch result
batch_result = BatchAnalysisResult(total_images = total_images,
processed = len(results),
failed = failed,
results = results,
total_processing_time = total_time,
)
# Calculate summary statistics
batch_result.summary = self._calculate_summary(results = results,
total = total_images,
)
logger.info(f"Batch processing complete: {len(results)}/{total_images} successful, {failed} failed in {total_time:.2f}s")
return batch_result
def _process_parallel(self, image_files: List[Dict], on_progress: Callable[[int, int, str], None] | None = None) -> Tuple[List[AnalysisResult], int]:
"""
Process images in parallel using ThreadPoolExecutor
Arguments:
----------
image_files { list } : List of image file dictionaries
on_progress { Callablel } : Optional callback invoked after each image is processed
Returns:
--------
{ tuple } : (results_list, failed_count)
"""
results = list()
failed = 0
logger.debug(f"Using parallel processing with {self.max_workers} workers")
with ThreadPoolExecutor(max_workers = self.max_workers) as executor:
# Submit all tasks
future_to_file = {executor.submit(self.process_single,
image['path'],
image['filename'],
image['size'],
): image for image in image_files
}
# Collect results as they complete
completed = 0
for future in as_completed(future_to_file):
completed += 1
image = future_to_file[future]
if on_progress:
on_progress(completed, len(image_files), image["filename"])
try:
result = future.result(timeout = settings.PROCESSING_TIMEOUT)
if result:
results.append(result)
logger.debug(f"✓ Completed: {image['filename']}")
else:
failed += 1
logger.warning(f"✗ Failed: {image['filename']} (returned None)")
except TimeoutError:
failed += 1
logger.error(f"✗ Timeout: {image['filename']} (exceeded {settings.PROCESSING_TIMEOUT}s)")
except Exception as e:
failed += 1
logger.error(f"✗ Error: {image['filename']} - {e}")
return results, failed
def _process_sequential(self, image_files: List[Dict], on_progress: Callable[[int, int, str], None] | None = None) -> Tuple[List[AnalysisResult], int]:
"""
Process images sequentially (fallback or single image)
Arguments:
----------
image_files { list } : List of image file dictionaries
on_progress { Callabel } : Optional callback invoked after each image is processed
Returns:
--------
{ tuple } : (results_list, failed_count)
"""
results = list()
failed = 0
logger.debug("Using sequential processing")
for idx, image in enumerate(image_files, 1):
try:
if on_progress:
on_progress(idx, len(image_files), image["filename"])
result = self.process_single(image_path = image['path'],
filename = image['filename'],
image_size = image['size'],
)
if result:
results.append(result)
logger.debug(f"✓ Completed: {image['filename']}")
else:
failed += 1
logger.warning(f"✗ Failed: {image['filename']} (returned None)")
except Exception as e:
failed += 1
logger.error(f"✗ Error: {image['filename']} - {e}")
return results, failed
def process_single(self, image_path: Path, filename: str, image_size: Tuple[int, int]) -> AnalysisResult:
"""
Process single image (called by both parallel and sequential)
Arguments:
----------
image_path { Path } : Path to image file
filename { str } : Original filename
image_size { tuple } : (width, height)
Returns:
--------
{ AnalysisResult } : Analysis result or None on error
"""
try:
return self.aggregator.analyze_image(image_path = image_path,
filename = filename,
image_size = image_size,
)
except Exception as e:
logger.error(f"Failed to process {filename}: {e}", exc_info = True)
return None
def _calculate_summary(self, results: List[AnalysisResult], total: int) -> Dict[str, int]:
"""
Calculate summary statistics from results
Arguments:
----------
results { list } : List of analysis results
total { int } : Total number of images
Returns:
--------
{ dict } : Summary statistics
"""
# Calculate processing stats
likely_authentic = sum(1 for r in results if (r.status == DetectionStatus.LIKELY_AUTHENTIC))
review_required = sum(1 for r in results if (r.status == DetectionStatus.REVIEW_REQUIRED))
processed = len(results)
failed = total - processed
success_rate = int((processed / total * 100) if (total > 0) else 0)
# Calculate average scores
avg_score = sum(r.overall_score for r in results) / len(results) if results else 0.0
avg_confidence = sum(r.confidence for r in results) / len(results) if results else 0
avg_proc_time = sum(r.processing_time for r in results) / len(results) if results else 0.0
return {"likely_authentic" : likely_authentic,
"review_required" : review_required,
"success_rate" : success_rate,
"processed" : processed,
"failed" : failed,
"avg_score" : round(avg_score, 3),
"avg_confidence" : int(avg_confidence),
"avg_proc_time" : round(avg_proc_time, 2),
}
def _create_empty_batch_result(self) -> BatchAnalysisResult:
"""
Create empty batch result for edge cases
Returns:
--------
{ BatchAnalysisResult } : Empty batch result
"""
return BatchAnalysisResult(total_images = 0,
processed = 0,
failed = 0,
results = [],
summary = {"likely_authentic" : 0,
"review_required" : 0,
"success_rate" : 0,
},
total_processing_time = 0.0,
)
|