[
https://issues.apache.org/jira/browse/SPARK-5981?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joseph K. Bradley updated SPARK-5981:
-------------------------------------
Description:
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.
was:
Many Python ML models and transformers use JavaModelWrapper to call methods in
the JVM, such as predict() and transform(). It is common to write:
{code}
data.map(lambda features: model.predict(features))
{code}
This fails because JavaModelWrapper.call uses the SparkContext (within the
transformation).
Note: It is possible to do a workaround using batch predict if
models/transformers support it:
{code}
model.predict(data)
{code}
However, this is still a major problem.
> 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]
