Streamlit OCR Document Processing Application

This application provides a web-based interface for OCR document processing, featuring a deduction form splitter and an invoice extractor. It is built with Streamlit and utilizes a Hugging Face transformer model for OCR, all containerized with Docker for easy deployment.

Installation and Setup

This guide assumes you have Docker and Docker Compose installed on your system.

1. Clone the Repository

If you haven't already, clone the repository to your local machine:

git clone https://github.com/jasoisjaso/ocrdeep.git
cd ocrdeep

2. Build the Docker Image

Navigate to the root directory of the cloned repository (where docker-compose.yml and Dockerfile are located) and build the Docker image:

docker-compose build

This process may take some time as it downloads the necessary base images and installs all Python dependencies, including the OCR model.

3. Run the Docker Containers

Once the image is built, you can start the application containers:

docker-compose up

If you want to run it in detached mode (in the background), use:

docker-compose up -d

4. Access the Application

Open your web browser and navigate to the following address:

http://localhost:8501

The Streamlit OCR application should now be running and accessible.