tag	758828d05f8756c19562182799197d3b01f9d30c
tagger	Kenneth Chan <kenneth@prediction.io>	Wed May 06 19:11:16 2015 -0700
object	6bc063dd786d98e580f0d66a7aa6a193a95873e1

commit	6bc063dd786d98e580f0d66a7aa6a193a95873e1	[log] [tgz]
author	Kenneth Chan <kenneth@prediction.io>	Wed May 06 19:11:04 2015 -0700
committer	Kenneth Chan <kenneth@prediction.io>	Wed May 06 19:11:04 2015 -0700
tree	e930b355fb7b1a3ed99959e32be36617678c4d80
parent	2357bb2a93349cd91203bef1a15e2048c2276762 [diff]
parent	dcfd600e27edf6f85b6e487436bf0b3038b47bfa [diff]

tree: e930b355fb7b1a3ed99959e32be36617678c4d80

README.md

Similar Product Template

Documentation

Please refer to http://docs.prediction.io/templates/similarproduct/quickstart/

Versions

v0.3.1

Add CooccurrenceAlgorithm. To use this algorithm, override engine.json by engine-cooccurrence.json, or specify --variant engine-cooccurrence.json parameter for both pio train and deploy

v0.3.0

update for PredictionIO 0.9.2, including:
- use new PEventStore API
- use appName in DataSource parameter

v0.2.0

update build.sbt and template.json for PredictionIO 0.9.2

v0.1.3

cache mllibRatings RDD in algorithm train() because it is used at multiple places (non-empty data check and ALS)

v0.1.2

update for PredictionIO 0.9.0

v0.1.1

Persist RDD to memory (.cache()) in DataSource for better performance
Use local model for faster serving.

v0.1.0

initial version

Development Notes

import sample data

$ python data/import_eventserver.py --access_key <your_access_key>

sample query

normal:

curl -H "Content-Type: application/json" \
-d '{ "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"], "num": 10}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}

curl -H "Content-Type: application/json" \
-d '{
  "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"],
  "num": 10,
  "categories" : ["c4", "c3"]
}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}

curl -H "Content-Type: application/json" \
-d '{
  "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"],
  "num": 10,
  "whiteList": ["i21", "i26", "i40"]
}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}

curl -H "Content-Type: application/json" \
-d '{
  "items": ["i1", "i3", "i10", "i2", "i5", "i31", "i9"],
  "num": 10,
  "blackList": ["i21", "i26", "i40"]
}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}

unknown item:

curl -H "Content-Type: application/json" \
-d '{ "items": ["unk1", "i3", "i10", "i2", "i5", "i31", "i9"], "num": 10}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}

all unknown items:

curl -H "Content-Type: application/json" \
-d '{ "items": ["unk1", "unk2", "unk3", "unk4"], "num": 10}' \
http://localhost:8000/queries.json \
-w %{time_connect}:%{time_starttransfer}:%{time_total}