[jira] [Commented] (SPARK-10884) Support prediction on single instance for regression and classification related models
[ https://issues.apache.org/jira/browse/SPARK-10884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16348094#comment-16348094 ] Ming Ma commented on SPARK-10884: - Cool. Could we get it in for 2.3? This will bring Spark one step closer providing real-time prediction and getting it to 2.3 will make it available to more applications sooner. Also the patch looks pretty straightforward and the risk seems pretty low. > Support prediction on single instance for regression and classification > related models > -- > > Key: SPARK-10884 > URL: https://issues.apache.org/jira/browse/SPARK-10884 > Project: Spark > Issue Type: Sub-task > Components: ML >Reporter: Yanbo Liang >Assignee: Yanbo Liang >Priority: Major > Labels: 2.2.0 > > Support prediction on single instance for regression and classification > related models (i.e., PredictionModel, ClassificationModel and their sub > classes). > Add corresponding test cases. > See parent issue for more details. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-10884) Support prediction on single instance for regression and classification related models
[ https://issues.apache.org/jira/browse/SPARK-10884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16347881#comment-16347881 ] Ming Ma commented on SPARK-10884: - Thanks [~WeichenXu123] and [~yanboliang]. While the long-term goal is to support "Pipeline for single instance" functionality, this specific patch is still quite useful. Any chance we can get it into the master branch soon? > Support prediction on single instance for regression and classification > related models > -- > > Key: SPARK-10884 > URL: https://issues.apache.org/jira/browse/SPARK-10884 > Project: Spark > Issue Type: Sub-task > Components: ML >Reporter: Yanbo Liang >Assignee: Yanbo Liang >Priority: Major > Labels: 2.2.0 > > Support prediction on single instance for regression and classification > related models (i.e., PredictionModel, ClassificationModel and their sub > classes). > Add corresponding test cases. > See parent issue for more details. -- This message was sent by Atlassian JIRA (v7.6.3#76005) - To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org
[jira] [Commented] (SPARK-15573) Backwards-compatible persistence for spark.ml
[ https://issues.apache.org/jira/browse/SPARK-15573?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16269140#comment-16269140 ] Ming Ma commented on SPARK-15573: - This is probably off topic as it isn't related to Spark version change, but the ability to evolve features and models regularly seems quite important. Can anyone confirm if how the scenario mentioned in http://apache-spark-user-list.1001560.n3.nabble.com/Spark-ML-Compatibility-between-features-and-models-td30100.html is currently handled? Thanks. > Backwards-compatible persistence for spark.ml > - > > Key: SPARK-15573 > URL: https://issues.apache.org/jira/browse/SPARK-15573 > Project: Spark > Issue Type: Improvement > Components: ML >Reporter: Joseph K. Bradley > > This JIRA is for imposing backwards-compatible persistence for the > DataFrames-based API for MLlib. I.e., we want to be able to load models > saved in previous versions of Spark. We will not require loading models > saved in later versions of Spark. > This requires: > * Putting unit tests in place to check loading models from previous versions > * Notifying all committers active on MLlib to be aware of this requirement in > the future > The unit tests could be written as in spark.mllib, where we essentially > copied and pasted the save() code every time it changed. This happens > rarely, so it should be acceptable, though other designs are fine. > Subtasks of this JIRA should cover checking and adding tests for existing > cases, such as KMeansModel (whose format changed between 1.6 and 2.0). -- 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
