INFO:     Started server process [14748]
INFO:     Waiting for application startup.
2026-06-23 10:41:09,401 \u2502 INFO    \u2502 Database initialized (with WAL mode + persistent connection and tasks table).
2026-06-23 10:42:20,150 \u2502 INFO    \u2502 Router: framed-manuscript model loaded (logic/philosophy)
2026-06-23 10:42:24,862 \u2502 INFO    \u2502 Router: framed-manuscript segmentation model loaded (logic/philosophy)
2026-06-23 10:42:24,887 \u2502 INFO    \u2502 Kraken v3 on GPU: NVIDIA GeForce RTX 4050 Laptop GPU (6.0GB), FP16=True
2026-06-23 10:42:24,894 \u2502 INFO    \u2502 TaskManager initialized with 3 max concurrent tasks (SQLite backed)
2026-06-23 10:42:24,900 \u2502 INFO    \u2502 Task Manager initialized with 3 workers
2026-06-23 10:42:24,906 \u2502 INFO    \u2502 \U0001f680 GPU detected: NVIDIA GeForce RTX 4050 Laptop GPU (6.0 GB)
2026-06-23 10:42:24,910 \u2502 INFO    \u2502    FP16 Mixed Precision: ON
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8001 (Press CTRL+C to quit)
2026-06-23 10:55:53,152 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 10:55:54,895 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 10:55:57,374 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:46:19,792 \u2502 INFO    \u2502 LMDecoder loaded:  (order=1, alpha=0.0, beam=6)
2026-06-23 11:46:37,997 \u2502 INFO    \u2502 HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
2026-06-23 11:49:00,842 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:49:02,549 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:50:08,759 \u2502 INFO    \u2502 HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
2026-06-23 11:52:27,653 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:52:30,012 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:53:20,172 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:53:20,445 \u2502 INFO    \u2502 \u23f1\ufe0f Phase 1 (preprocess-light): 0.23s
2026-06-23 11:53:20,481 \u2502 INFO    \u2502 \u23f1\ufe0f Phase 1b (preprocess, enhance=False): 0.03s
2026-06-23 11:53:20,484 \u2502 INFO    \u2502 \u23f1\ufe0f Phase 1c (binarize=False): 0.00s
2026-06-23 11:53:20,765 \u2502 INFO    \u2502 Router: requested=auto segmentation=logic recognition=logic reason=dense_red_layout:ink=0.147,rows=9,red=0.0777,content=1.00x1.00 frame=None
2026-06-23 11:53:20,767 \u2502 INFO    \u2502 Using logic/philosophy neural segmentation model (BLLA) to generate baselines compatible with HTR...
2026-06-23 11:53:20,804 \u2502 INFO    \u2502 Segmenting <PIL.Image.Image image mode=RGB size=940x1280 at 0x15644C3A9B0>
2026-06-23 11:53:38,032 \u2502 INFO    \u2502 Vectorizing regions
2026-06-23 11:53:40,762 \u2502 INFO    \u2502 Vectorizing baselines
2026-06-23 11:56:42,510 \u2502 INFO    \u2502 Compute reading order on 46 lines in lr direction
2026-06-23 11:56:42,520 \u2502 INFO    \u2502 Perform topological sort on partially ordered lines
2026-06-23 11:56:42,524 \u2502 INFO    \u2502 Compute reading order on 1 lines in lr direction
2026-06-23 11:56:42,526 \u2502 INFO    \u2502 Perform topological sort on partially ordered lines
2026-06-23 11:56:42,592 \u2502 INFO    \u2502 \u23f1\ufe0f Phase 2 (segment/neural): 201.83s
2026-06-23 11:56:42,600 \u2502 INFO    \u2502 Boundary expansion skipped for logic/philosophy route
2026-06-23 11:56:42,601 \u2502 INFO    \u2502 Router: specialized recognition model selected
C:\Users\aiden\Desktop\\u0645\u0633\u0627\u0628\u0642\u0629 \u0648\u0637\u0646\u064a\u0629\ocr-fpr-phonex-team-main\backend\.venv\lib\site-packages\kraken\lib\segmentation.py:1098: RuntimeWarning: divide by zero encountered in divide
  t1 = (v[:, 0] * u[:, 1] - v[:, 1] * u[:, 0]) / (z + np.finfo(float).eps)
C:\Users\aiden\Desktop\\u0645\u0633\u0627\u0628\u0642\u0629 \u0648\u0637\u0646\u064a\u0629\ocr-fpr-phonex-team-main\backend\.venv\lib\site-packages\kraken\lib\segmentation.py:1099: RuntimeWarning: divide by zero encountered in divide
  t2 = np.dot(u, w) / (z + np.finfo(float).eps)
2026-06-23 11:57:14,805 \u2502 INFO    \u2502 \u23f1\ufe0f Phase 3 (recognize): 32.20s
2026-06-23 11:57:18,719 \u2502 INFO    \u2502 LMDecoder loaded: C:\Users\aiden\Desktop\\u0645\u0633\u0627\u0628\u0642\u0629 \u0648\u0637\u0646\u064a\u0629\ocr-fpr-phonex-team-main\\u0646\u0645\u0627\u0630\u062c\char8_trainval.lm (order=8, alpha=0.5, beam=10)
2026-06-23 11:57:18,725 \u2502 INFO    \u2502 Gated LM decoder loaded
2026-06-23 11:57:20,215 \u2502 INFO    \u2502 Adaptive merge parameters: vertical_threshold=20px, max_horizontal_gap=30px
2026-06-23 11:57:20,218 \u2502 INFO    \u2502 Region classify: {'main': 39}
2026-06-23 11:57:20,451 \u2502 INFO    \u2502 PostProcessor: 39 results, 0 corrections applied
2026-06-23 11:57:20,495 \u2502 INFO    \u2502 \u23f1\ufe0f Kraken TOTAL: 240.28s | 45 raw \u2192 39 merged \u2192 39 final
2026-06-23 11:57:21,234 \u2502 INFO    \u2502 \u23f1\ufe0f Total stream_mode1: 241.05s
2026-06-23 11:57:32,527 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:57:34,734 \u2502 WARNING \u2502 Skipping data after last boundary
2026-06-23 11:58:23,842 \u2502 INFO    \u2502 HTTP Request: POST https://api.openai.com/v1/chat/completions "HTTP/1.1 200 OK"
2026-06-23 13:01:09,835 \u2502 INFO    \u2502 Catalog loaded: 35632 works from works_index.json