Bootstrap your first Streamlit web application in seconds using this docker container with dependencies like pandas, numpy, matplotlib, plotly... already installed.
git clone https://github.com/bgeneto/streamlit-app.gitFirst you need to build the "streamlit-base" docker image. The proporse of this image is to serve as a base image to all your deployed streamlit app containers.
cd streamlit-app/streamlit-base
./buildIf you don't have a localhost docker registry running, then issue the following command instead to avoid uploading this image to a non-existing local registry:
./build --no-regYou can also edit files packages.txt and requirements.txt to add the packages and
softwares that are required in your container.
Now we are ready to build the main app image/container.
But first edit the .env file in the root dir and then call docker compose.
cd ..
docker compose up
Note: The provided
docker-compose.ymlare ready to use but it depends on autoheal to monitor your streamlit container status. Just remember to run the autoheal container beforehand withAUTOHEAL_CONTAINER_LABEL=autoheal.