Comparison to other implementations

[benmccann]

I found several other implementations of FM. I'm curious about the design choices made in the Predicteworks implementation differ from these other implementations

https://github.com/zhengruifeng/spark-libFM https://github.com/blebreton/spark-FM-parallelSGD https://github.com/witgo/zen/tree/master/ml/src/main/scala/com/github/cloudml/zen/ml/recommendation

[eherrerosj]

I'm also interested in any comments should someone has tried different approaches using FM in Spark.