REST web service for the true real-time scoring (<1 ms) of R, Scikit-Learn and Apache Spark models
869
Openscoring application for the Docker distributed applications platform
[Openscoring] (https://github.com/jpmml/openscoring) provides REST API for publishing and evaluating predictive models:
Prerequisites:
GitHub repository [jpmml/openscoring-docker] (https://github.com/jpmml/openscoring-docker) contains a Dockerfile for Openscoring command-line server application.
Building the latest Openscoring application image from the HEAD revision:
sudo docker build -t jpmml/openscoring:latest github.com/jpmml/openscoring-docker
Additionally, this GitHub repository is tracked by Docker Hub repository [jpmml/openscoring] (https://registry.hub.docker.com/u/jpmml/openscoring/) using the "Automated Builds" mechanism.
Pulling a stable Openscoring application image:
sudo docker pull jpmml/openscoring:1.2.2
Running the image in the host networking mode:
sudo docker run --net="host" jpmml/openscoring:latest
The container shares host's network stack. It is possible to use privileged HTTP methods PUT and DELETE for deploying and undeploying models, respectively.
Running the image in the bridge (default) networking mode:
sudo docker run --net="bridge" -p 8080:8080 -v /path/to/pmml:/openscoring/pmml jpmml/openscoring:latest --model-dir /openscoring/pmml
The container uses Docker's default network setup, which is separate from host's network stack. It is impossible to use privileged HTTP methods. The only option for deploying and undeploying models is via the model auto-deployment directory /openscoring/pmml. This directory is mapped to host's filesystem directory /path/to/pmml using the data volume mechanism.
Openscoring is dual-licensed under the [GNU Affero General Public License (AGPL) version 3.0] (http://www.gnu.org/licenses/agpl-3.0.html) and a commercial license.
Please contact [[email protected]] (mailto:[email protected])
Content type
Image
Digest
Size
259.9 MB
Last updated
over 8 years ago
docker pull ggirou/openscoring