Xiangrui Meng created SPARK-5890:
------------------------------------
Summary: Add FeatureDiscretizer
Key: SPARK-5890
URL: https://issues.apache.org/jira/browse/SPARK-5890
Project: Spark
Issue Type: Sub-task
Components: ML
Reporter: Xiangrui Meng
A `FeatureDiscretizer` takes a column with continuous features and outputs a
column with binned categorical features.
{code}
val fd = new FeatureDiscretizer()
.setInputCol("age")
.setNumBins(32)
.setOutputCol("ageBins")
{code}
This should an automatic feature discretizer, which uses a simple algorithm
like approximate quantiles to discretize features. It should set the ML
attribute correctly in the output column.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
