Burak Yavuz created SPARK-7492:
----------------------------------
Summary: Convert LocalDataFrame to LocalMatrix
Key: SPARK-7492
URL: https://issues.apache.org/jira/browse/SPARK-7492
Project: Spark
Issue Type: New Feature
Components: MLlib, SQL
Reporter: Burak Yavuz
Having a method like,
{code: scala}
Matrices.fromDataFrame(df)
{code}
would provide users the ability to perform feature selection with DataFrames.
Users will be able to chain operations like below:
{code: scala}
import org.apache.spark.mllib.linalg.Matrices
import org.apache.spark.mllib.stat.Statistics
import org.apache.spark.sql.DataFrame
val df = ... // the DataFrame
val contingencyTable = df.stat.crosstab(col1, col2)
val ct = Matrices.fromDataFrame(contingencyTable)
val result: ChiSqTestResult = Statistics.chiSqTest(ct)
{code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
