Skip to content

Latest commit

 

History

History
61 lines (37 loc) · 2.31 KB

README.md

File metadata and controls

61 lines (37 loc) · 2.31 KB
Raw

feedzai Open Scoring Server (FOS) - Weka Scoring and Training module

fos-weka is a fos-core supporting Weka classifiers

Build Status

CloudbeesDevCloud

Why FOS

There are pretty good machine learning training and scoring frameworks/libraries out there, but they don't provide the following benefits:

  1. Common API: fos provides a common abstraction for model attributes, model training and model scoring. Using a Weka based classifier will use have exactly the same API as using a R based classifier.
  2. Scoring & Training as a remote service: Training and scoring can be farmed to dedicated servers in the network enabling both vertical and horizontal scaling.
  3. Import and Export models: A model could be trained in a development box and imported seamlessly into a remote server
  4. Scalable and low latency scoring: Marshalling and Unmarshalling scoring requests/responses can be responsible for a significant amount of overhead. Along with the slow RMI based interface, fos also supports scoring using Kryo.

Compiling fos-weka

You need:

  1. Java SDK: Java 7
  2. Maven: Tested with maven 3.0.X
  3. fos-core
  4. Access to maven central repo (or a local proxy)

After both the Java SDK and Maven are installed run the following command

mvn clean install

This should compile fos-core, ran all the tests and install all modules into your local maven repo.

Running FOS

FAQ

Weka is great but i need to support another library. How can I do it?

fos-core does not provide any concrete implementation. You may want to peek at fos-impl-dummy module in fos-core. After you've familiarized with the API (Manager, ManagerFactory and Scorer) you can take a peek at a real implementation, say fos-weka.