MLI
MLI copied to clipboard
An API for Distributed Machine Learning
Hullo, This pull request adds a corresponding `sc.stop()` for each `val sc = new SparkContext(...)` to avoid `address already in use` errors that come up running `./sbt/sbt test` against the...
Tree
Decision Tree algorithm implemented on top of Spark RDD. Key features: - Supports both classification and regressions - Supports gini, entropy and variance for information gain calculation - Supports calculating...
Apparently Scala 2.10 has added method with the same name `toVector` to `IndexedSeq` class, which cause compilation problems when trying MLI with latest Spark 0.9