The CoughOverFlow Analysis Engine binaries for amd64, arm64.
To run the analysis engine:
./overflowengine --input image_location --output output_location.txt
- The
--input
parameter is the absolute or relative location of the image (only jpg) to be analysed. - The
--output
parameter is the absolute or relative location of a text file to hold the analysis result.
To install the analysis engine inside a Docker container:
dpkg --print-architecture | grep -q "amd64" && export ARCH="amd64" || export ARCH="arm64" && wget https://github.com/CSSE6400/CoughOverflow-Engine/releases/download/v1.0/overflowengine-${ARCH} -O overflowengine && chmod +x overflowengine
The provided Dockerfile
installs the analysis engine into a minimal Ubuntu image with the libraries required to run the analysis engine.
It then copies the sample JPG images into the container so you can experiment with the analysis engine.
- The
sample-images
directory contains a set of prepared samples to use for testing. The file names indicate the expected result from the analysis engine - The
sample-base64
directory contains base64 encodings of the sample JPGs in thesample-images
directory.
You can clone this repository, build the container and run it.
docker build -t engine .
docker run -it --rm engine
This allows you to interact with the container through your terminal. You can then execute the analysis engine manually to experiment with how it works.
For the assignment, you will need to run the analysis engine on the JPG images you receive via a POST request.