Recommendation as a Service
This docker image contains recommendation engine based on https://prediction.io and Universal Recommender created by Pat Ferrel here https://github.com/pferrel/template-scala-parallel-universal-recommendation .
How to use
docker pull goliasz/raas-micro:1.2
Once pulled just create a profile in PubNub. Find your PubKey and SubKey. Imagine your channel name and use this information for running the container.
docker run -it --hostname reco0 --name reco0 -e "PN_PUBKEY=pub-c-1113-demo-3" -e "PN_SUBKEY=sub-c-1f1a-demo" -e "PN_REQ_CHANNEL=Channel-my-demo0" -e "PN_RES_CHANNEL=Channel-my-demo0" goliasz/raas-micro:1.2 /MyEngine/autostart.sh
Once container is started it is ready for training and serving recommendations.
Some introduction here http://mlaas-adventures.blogspot.com/2016/01/recommendation-as-microservice.html
Some technical information here http://raas.recomme.eu/