[ https://issues.apache.org/jira/browse/SPARK-13490?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Yanbo Liang updated SPARK-13490: -------------------------------- Description: Like SPARK-13132 for LogisticRegression, LinearRegression with L1 regularization should also cache the value of the standardization rather than re-fetching it from the ParamMap for every OWLQN iteration. (was: Like SPARK-13132 for LogisticRegression, when LinearRegression with L1 regularization, the inner functor passed to the quasi-newton optimizer in {{org.apache.spark.ml.regression.LinearRegression#train}} makes repeated calls to {{$(standardization)}}. This ultimately involves repeated string interpolation triggered by {{org.apache.spark.ml.param.Param#hashCode}}. We should cache the value of the standardization rather than re-fetching it from the ParamMap for every iteration.) > ML LinearRegression should cache standardization param value > ------------------------------------------------------------ > > Key: SPARK-13490 > URL: https://issues.apache.org/jira/browse/SPARK-13490 > Project: Spark > Issue Type: Improvement > Components: ML > Reporter: Yanbo Liang > Priority: Minor > > Like SPARK-13132 for LogisticRegression, LinearRegression with L1 > regularization should also cache the value of the standardization rather than > re-fetching it from the ParamMap for every OWLQN iteration. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org