predictionio-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From TJ <>
Subject engine.json for Universal recommender.
Date Thu, 15 Dec 2016 11:42:49 GMT
I am a complete novice with this and really need help configuring the engine.json for my app.

This is what i know…..

I have these events.

New user sign-up with one parameter - “social-status”

“Rate item” event
“Mark as favourite event”
“Buy” event

I have 36000 recipes (food) that have the following parameters. 

id, main ingredient, diet type, cooking time, meal type. 

These last 4 are user as a filter when making queries to the engine. 

So my question is first, how do I construct the engine.json . I have read a lot of documentation
but I am still unsure how to construct it. 

This is what I have so far but there json is invalid because of duplicate keys. (“name”)

  "comment":" This config file uses default settings for all but the required values see
for docs",
  "id": "iqmotor",
  "description": "Default settings",
  "engineFactory": "org.template.RecommendationEngine",
  "datasource": {
    "params" : {
      "name": "iqchef-motor",
      "appName": "iqmotor",
      "eventNames": ["rate", "buy", "favourite"],
  "sparkConf": {
    "spark.serializer": "org.apache.spark.serializer.KryoSerializer",
    "spark.kryo.registrator": "",
    "spark.kryo.referenceTracking": "false",
    "spark.kryoserializer.buffer": "300m",
    "": "true"
  "algorithms": [
      "comment": "simplest setup where all values are default, popularity based backfill,
must add eventsNames",
      "name": "ur",
      "params": {
        "appName": "iqmotor",
        "indexName": "urindex",
        "typeName": "items",
        "comment": "must have data for the first event or the model will not build, other
events are optional",
        "eventNames": ["rate","buy","favourite"],
        "num" : 200,
        "fields" : [

                        "name" : "mainingredient",
                        "values" : ["9","10","12","13","7","11","5","4","15"],

                        "name" : "diettype",
                        "values" : ["1","25","20","18","17","16","9","7","3"],

                        "name" : "cookingtime",
                        "values" : ["fast" , "slow" ],

                        "name" : "mealtype",
                        "values" : ["76","48","26","17","29","47","3","2","14"],



View raw message