xtracture 0.4.0

Xtracture is an open source library designed to efficiently extract arbitrary elements from documents.

Xtracture: Open Source Document Content Extractuion Library

Xtracture is an open source library designed to efficiently extract arbitrary elements from documents.

Features

• Natural language rule creation using LLMs
• Switchable OCR engines for optimized perfomance and accuracy

prerequirements

• OpenAI API Key (for LLM rule creation)

Installation

pip install -U xtracture

Usage

Use Google Cloud Vision API

Google CLoud Vision Credentials must be correctly configured.

see examples/google_cloud_vision_example.py.

Use Tesseract

Tesseract must be installed beforehand.

see examples/tesseract_example.py.

Use only GPT Extractor

You can input OCR-processed text file. see examples/lambda_example.py.

License

Xtracture is released under the MIT License.