feat: refactor summarizer and PDF extraction pipeline

- Split summarizer into summary_generator and summary_persister modules
- Refactor pdf_image_extractor to two-phase pipeline with PicoDet layout detection
- Add layout_detector service for PicoDet-S_layout_3cls integration
- Add exceptions module with ConflictError and NotFoundError
- Improve admin dashboard with better statistics and task management
- Add design review document with system optimization suggestions
- Add new tests for crawler, pdf_downloader, pipeline, and summary_utils
- Update dependencies and configuration
- Clean up dead code and improve error handling
This commit is contained in:
2026-06-13 13:16:47 +08:00
parent e2f0e1a8be
commit 21f16e6756
43 changed files with 3304 additions and 1494 deletions
+5
View File
@@ -46,3 +46,8 @@ EMBED_API_BASE=https://api.siliconflow.cn/v1/embeddings
EMBED_API_KEY=your_api_key_here
EMBED_MODEL=Qwen/Qwen3-Embedding-4B
EMBED_DIMENSIONS=2560
# ─── 布局检测 ─────────────────────────────
# ONNX 模型路径(首次运行前执行 scripts/export_picodet_onnx.py 导出)
# LAYOUT_MODEL_PATH=data/models/picodet_layout_3cls.onnx
# LAYOUT_THRESHOLD=0.5