[
https://issues.apache.org/jira/browse/SPARK-5981?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14347237#comment-14347237
]
Joseph K. Bradley commented on SPARK-5981:
------------------------------------------
When predict() is called on a single vector or an RDD, it's being called on the
driver, has access to the SparkContext (which is required by
JavaModelWrapper.call). But when predict() is called within an RDD.map, it's
being called on workers, which don't have the SparkContext.
> pyspark ML models should support predict/transform on vector within map
> -----------------------------------------------------------------------
>
> Key: SPARK-5981
> URL: https://issues.apache.org/jira/browse/SPARK-5981
> Project: Spark
> Issue Type: Improvement
> Components: MLlib, PySpark
> Affects Versions: 1.3.0
> Reporter: Joseph K. Bradley
>
> Currently, most Python models only have limited support for single-vector
> prediction.
> E.g., one can call {code}model.predict(myFeatureVector){code} for a single
> instance, but that fails within a map for Python ML models and transformers
> which use JavaModelWrapper:
> {code}
> data.map(lambda features: model.predict(features))
> {code}
> This fails because JavaModelWrapper.call uses the SparkContext (within the
> transformation). (It works for linear models, which do prediction within
> Python.)
> Supporting prediction within a map would require storing the model and doing
> prediction/transformation within Python.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
