[ https://issues.apache.org/jira/browse/SPARK-8515?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16206770#comment-16206770 ]
Liang-Chi Hsieh edited comment on SPARK-8515 at 10/16/17 11:28 PM: ------------------------------------------------------------------- I'm not sure if SPARK-2008 is related to metadata in ML? I'm agreed with [~timhunter], the design and the implementation is basically come out to address sparsity and efficiency issues of current ML attribute. So I assume it goes to satisfy most the user cases that the current ML attribute enables. If we are sure most of features in current ML attribute are redundant, we have get rid of it. was (Author: viirya): I'm not sure if SPARK-2008 is related to metadata in ML? I'm agreed with [~timhunter], the design and the implementation is basically come out to address sparsity and efficiency issues of current ML attribute. So I assume it goes to satisfy most the user cases that the current attribute ML enables. If we are sure we don't need most of features, we have get rid of it. > Improve ML attribute API > ------------------------ > > Key: SPARK-8515 > URL: https://issues.apache.org/jira/browse/SPARK-8515 > Project: Spark > Issue Type: Improvement > Components: ML > Affects Versions: 1.4.0 > Reporter: Xiangrui Meng > Labels: advanced > Attachments: SPARK-8515.pdf > > > In 1.4.0, we introduced ML attribute API to embed feature/label attribute > info inside DataFrame's schema. However, the API is not very friendly to use. > We should re-visit this API and see how we can improve it. -- This message was sent by Atlassian JIRA (v6.4.14#64029) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org