Skip to content

Pipeline API

Main pipeline class for document processing.

Overview

The Pipeline class orchestrates the entire 8-stage document processing workflow.

from pathlib import Path
from pipeline import Pipeline

# Create pipeline
pipeline = Pipeline(
    detector_name="doclayout-yolo",
    sorter_name="mineru-xycut",
    backend="gemini",
    model="gemini-2.5-flash"
)

# Process PDF
result = pipeline.process_single_pdf(Path("document.pdf"))

Key Methods

__init__

Initialize the pipeline with detector, sorter, and recognizer configurations.

process_single_pdf

Process a single PDF file and return results.

process_directory

Batch process all PDFs in a directory.

process_page

Process a single page (used internally).

Full API Reference

Detailed API reference coming soon. See Advanced Examples for usage patterns.