Copyright (c) General Electric Company, 2017. All rights reserved.
This is an example analytic for Rt 106. This analytic demonstrates how to use the rt106-algorithm-sdk around an ITK algorithm.
Two Docker images are used to package and deploy the analytic. The first is a dev
Docker image that is responsible for compiling the C++ code. The dev
contains a full build environment (compilers). Building the dev
image also compiles our algorithm code. After the dev
image is built, a call to docker run
emits a tarball containing the artifacts from the build. Here, The second Docker image ops
installs the aforementioned tarball into a runtime container. The two container strategy allows for the runtime ops
container to be small.
To build the dev
container
$ cd docker
$ docker build -t rt106-dev/rt106-wavelet-nuclei-segmentation --build-arg http_proxy=$http_proxy --build-arg https_proxy=$https_proxy --build-arg no_proxy=$no_proxy dev
To build the ops
container, we first need to move the artifact from the dev
container
into place to build the ops
container. If we run the dev
container just built, it will
emit the artifact as a tarball
$ cd docker
$ docker run rt106-dev/rt106-wavelet-nuclei-segmentation > ops/rt106-wavelet-nuclei-segmentation.tar.gz
To build the ops
container
$ cd docker
$ docker build -t rt106/rt106-wavelet-nuclei-segmentation --build-arg http_proxy=$http_proxy --build-arg https_proxy=$https_proxy --build-arg no_proxy=$no_proxy ops