About | Features | Technologies | Requirements | Starting | License | Author
This project is about document data extraction and document classification.
✔️ Document Data Extraction; ✔️ Document Classification;
The following tools were used in this project:
Before starting 🏁, you need to have Git or Docker installed.
# Clone this project
$ git clone https://github.com/human-ai2025/document_ocr_classification
# Access
$ cd document_ocr_classification
#
# For devlopment
docker compose -f build_run\deploy\docker-compose-dev.yaml down && docker build -f dev.Dockerfile -t pytorch_doc_ocr_classification_deploy:latest . && docker compose -f build_run\deploy\docker-compose-dev.yaml up
# For testing inferencing in bash mode
docker compose -f build_run\deploy\docker-compose-deploy-bash.yaml down && docker build -f deploy.Dockerfile -t pytorch_doc_ocr_classification_deploy:latest . && docker compose -f build_run\deploy\docker-compose-deploy-bash.yaml up -d
docker exec -it deploy-document-classification-1 bash
#For Inferencing (This is the part you need)
$ docker compose -f build_run\deploy\docker-compose-deploy.yaml down && docker build -f deploy.Dockerfile -t pytorch_doc_ocr_classification_deploy:latest . && docker compose -f build_run\deploy\docker-compose-deploy.yaml up
# Locally in windows/Mac machine most probably EOL error may come dur to the script file
#Change the EOL of startup.sh in build_run/deploy to overcome those issues
# Now visit http://localhost:8501/ in your browser
# NOTES
# For interactive bash include the folllowing lines in docker compose under your service
tty: true
stdin_open: true
command: sh
# To get the list versions only in pip
pip3 list --format=freeze > requirements.txt
# To run stramlit
streamlit run app.py --server.port=8501 --server.address=0.0.0.0
View @ http://localhost:8501/
# The server will initialize in the <>
This project is under license from MIT. For more details, see the LICENSE file.
Made with ❤️ by human-ai2025