alexcheng1982 / predictionio-sample

PredicitionIO sample

PredictionIO Sample

This is a quick start sample for using PredictionIO. The template used is template-java-parallel-ecommercerecommendation. PredictionIO runs in a Docker container.

Run the sample

Start container

Start Docker container using Docker Compose.

$ docker-compose build
$ docker-compose -p pio up -d

The container will create the app and build the engine. It may take some time to finish. Use docker logs to watch the container logs to check when it's finished.

$ docker logs pio_pio_1

Load sample data

Use Python script to load sample data. Access key can be found in container logs. pio_pio_1 is conatiner's name.

$ docker exec pio_pio_1 python data/import_eventserver.py --access_key <access key>

Train

Train the data.

$ docker exec pio_pio_1 pio train

Deploy the engine

Deploy the engine. The engine runs in container port 8000, which is mapped to host port 10080.

$ docker exec pio_pio_1 pio deploy

Run the query

Run the query on host machine using curl. https is required.

$ curl -k -H "Content-Type: application/json" \
-d '{
  "userEntityId" : "u1",
  "number" : 10 }' \
https://localhost:10800/queries.json