cv.T206.io

A computer vision app for T206.io (GitHub repo)

Made with:

• Python 2.7
• OpenCV
• HDF5
• Docker
• Hosted on Digital Ocean
• https://github.com/jdesilvio/py-opencv-hdf5-docker as a starting point for Docker environment
• https://github.com/realpython/flask-image-search as a starting point for the Flask app
• https://github.com/OrangeTux/minimal-docker-python-setup for NGINX support

Setup

# Import NGINX image during initial setup
docker import - jdesilvio/nginx < nginx/rootfs.tar

# Build
docker-compose build &&
docker-compose up -d &&
docker exec t206cv_app_1 mkdir app/dbdata &&
docker cp ~/Desktop/dbdata/data.h5 t206cv_app_1:app/dbdata/data.h5

# Re-build from scratch
docker-compose stop &&
docker-compose rm -f --all &&
docker-compose pull &&
docker-compose build --no-cache &&
docker-compose up -d --force-recreate --remove-orphans

If errors appear...

And it sounds like a memory issue:

• Create a swapfile
• It could be that there is a bunch of junk in your docker-machine. Check the size of a file called disk.vmdk. If it's large and you are ok with deleting all images and containers and starting from scratch, run: docker-machine rm default

---

Setup without Docker Compose (keeping for reference)

docker build --rm --build-arg ACCESS_KEY='access key' --build-arg SECRET_KEY='secret key' -t t206cv-data .
docker build --rm -t t206cv .
docker run -d -t -v /app/dbdata --name data t206cv-data
docker run -p 80:5000 --volumes-from data --name app t206cv