How opendatalab/pdf-extract-kit works
PDF-Extract-Kit is a modular toolkit of document-parsing models, with a sample PDF-to-Markdown application as its end-to-end workflow. That application renders pages, combines layout and formula detections with OCR and formula recognition, then merges spatial results into Markdown. Separately, users can run task-specific detection, recognition, OCR, and table-parsing tasks. The source excerpts show the PDF rendering, formula-detection, and result-merging links; some application-stage wiring and model internals are unsampled, so those links are omitted or marked unverified.
Main entry points: pdf2markdown.py PDF-to-Markdown · config_loader.py Config Loader · base_task.py Base Task · task.py Layout Task
Components 20
Task Runtime
- Config Loader
pdf_extract_kit/utils/config_loader.py - Task and Model Registry
pdf_extract_kit/registry/registry.py - Base Task
pdf_extract_kit/tasks/base_task.py - Layout Task
pdf_extract_kit/tasks/layout_detection/task.py - Formula Detection Task
pdf_extract_kit/tasks/formula_detection/task.py - Formula Recognition Task
pdf_extract_kit/tasks/formula_recognition/task.py - OCR Task
pdf_extract_kit/tasks/ocr/task.py - Table Parsing Task
pdf_extract_kit/tasks/table_parsing/task.py
Parsing Models
- Layout Models
pdf_extract_kit/tasks/layout_detection/models - YOLO Formula Detector
pdf_extract_kit/tasks/formula_detection/models/yolo.py - UniMERNet
pdf_extract_kit/tasks/formula_recognition/models/unimernet.py - PaddleOCR
pdf_extract_kit/tasks/ocr/models/paddle_ocr.py - StructEqTable
pdf_extract_kit/tasks/table_parsing/models/struct_eqtable.py
PDF to Markdown
- PDF-to-Markdown
project/pdf2markdown/scripts/pdf2markdown.py - Markdown Output
project/pdf2markdown/scripts/pdf2markdown.py
Document Utilities
- PDF Page Rendering
pdf_extract_kit/utils/data_preprocess.py - Image Datasets
pdf_extract_kit/dataset/dataset.py - Block and Span Merging
pdf_extract_kit/utils/merge_blocks_and_spans.py - Result Visualization
pdf_extract_kit/utils/visualization.py
Other components
- User
Connections 27
- User → Config Loader: provides config
pdf_extract_kit/utils/config_loader.py - Config Loader → Task and Model Registry: resolves classes
pdf_extract_kit/utils/config_loader.py - Config Loader → Layout Task: instantiates
pdf_extract_kit/utils/config_loader.py - Config Loader → Formula Detection Task: instantiates
pdf_extract_kit/utils/config_loader.py - Config Loader → Formula Recognition Task: instantiates
pdf_extract_kit/utils/config_loader.py - Config Loader → OCR Task: instantiates
pdf_extract_kit/utils/config_loader.py - Config Loader → Table Parsing Task: instantiates
pdf_extract_kit/utils/config_loader.py - Layout Task → Base Task: inherits
pdf_extract_kit/tasks/layout_detection/task.py - Formula Detection Task → Base Task: inherits
pdf_extract_kit/tasks/formula_detection/task.py - Formula Recognition Task → Base Task: inherits
pdf_extract_kit/tasks/formula_recognition/task.py - OCR Task → Base Task: inherits
pdf_extract_kit/tasks/ocr/task.py - Formula Detection Task → YOLO Formula Detector: calls predict
pdf_extract_kit/tasks/formula_detection/task.py - Formula Recognition Task → UniMERNet: uses model
pdf_extract_kit/tasks/formula_recognition/task.py - OCR Task → PaddleOCR: calls predict
pdf_extract_kit/tasks/ocr/task.py - Table Parsing Task → StructEqTable: uses model
pdf_extract_kit/tasks/table_parsing/task.py - Layout Task → Layout Models: uses model
pdf_extract_kit/tasks/layout_detection/task.py - User → PDF-to-Markdown: runs extraction
project/pdf2markdown/scripts/pdf2markdown.py - PDF-to-Markdown → PDF Page Rendering: renders pages
project/pdf2markdown/scripts/pdf2markdown.py - PDF-to-Markdown → YOLO Formula Detector: detects formulas
project/pdf2markdown/scripts/pdf2markdown.py - PDF-to-Markdown → Layout Models: requests layout
- PDF-to-Markdown → PaddleOCR: requests OCR
- PDF-to-Markdown → UniMERNet: recognizes formulas
- PDF-to-Markdown → Image Datasets: uses dataset
project/pdf2markdown/scripts/pdf2markdown.py - PDF-to-Markdown → Block and Span Merging: merges spans
project/pdf2markdown/scripts/pdf2markdown.py - PDF-to-Markdown → Markdown Output: writes Markdown
project/pdf2markdown/scripts/pdf2markdown.py - Layout Models → Result Visualization: visualizes results
pdf_extract_kit/tasks/layout_detection/models/layoutlmv3.py - YOLO Formula Detector → Image Datasets: loads batches
pdf_extract_kit/tasks/formula_detection/models/yolo.py