This is a flask app that presents a UI to help image annotators annotate human activities using vision and language models.
You can clone this repository. Before you begin, ensure all the packages are installed by running the following command in a virtual environment:
cd path/to/your/directory
python -m venv myvenv
source myenv/bin/activate
pip install -r requirements.txt
If you are running the app for the first time, you will need to initialize the database. You can do so by opening a python interpreter in your virtual environment. Run the following commands to initialize the database.
>>> from llm_app import app, db
>>> app.app_context().push()
>>> db.create_all()
If you encounter any problems with the database, try deleting the database at instance/site.db and run the above commands again.
You can run the app in your localhost by running the following command:
python3 run.py