EagleEye

Observing mice from the beginning till the end 🦅

Run the container

Prerequisites

To supply the app with data you have to provide a docker volume with the following layout:

• the root path must contain the directories "models" and "input".
• there is a file "/input/features-with-indicators.pkl" that contains the original training data.
• "/models" contains a set of NetCDF files. The names of the these files will appear in the mouse ID drop down menu.
• you can use the volume "bar-output" for testing. Please make sure to not overwrite any data on it!

Start the app

Foward a port from your port range to Docker host:

ssh username@iam-docker -NL {your-port}:localhost:{your-port}

Run the following command to start EagleEye on Docker host:

docker run -it --rm -v {my-volume}:/data -p {your-port}:8501 eagleeye

Navigate your browser to localhost:{your-port}.