You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spark.apache.org by harini <ph...@gmail.com> on 2016/12/07 23:00:23 UTC
modifications to ALS.scala
Hi all,I am trying to implement ALS with a slightly modified objective
function, which will require minor changes to fit -> train -> computeFactors
within ALS.scala
<https://github.com/apache/spark/blob/v1.6.2/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala>
- Is there a way to do this without having to build spark in its entirety?
--
View this message in context: http://apache-spark-developers-list.1001551.n3.nabble.com/modifications-to-ALS-scala-tp20167.html
Sent from the Apache Spark Developers List mailing list archive at Nabble.com.
Re: modifications to ALS.scala
Posted by Georg Heiler <ge...@gmail.com>.
You can write some code e.g. A custom estimator transformer in sparks
namespace.
http://stackoverflow.com/a/40785438/2587904 might help you get started.
Be aware that using private e.g. Spark internal api might be subjected to
change from release to release.
You definitely will require spark -mllib dependency.
Currently for my usage I was not required to build a separate version of
mllib.
harini <ph...@gmail.com> schrieb am Do. 8. Dez. 2016 um 00:23:
> I am new to development with spark, how do I do that? Can I write up a
> custom
> implementation under package org.apache.spark.ml.recommendation, and
> specify
> "spark-mllib" along with others as a library dependency?
>
>
>
> --
> View this message in context:
> http://apache-spark-developers-list.1001551.n3.nabble.com/modifications-to-ALS-scala-tp20167p20169.html
> Sent from the Apache Spark Developers List mailing list archive at
> Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe e-mail: dev-unsubscribe@spark.apache.org
>
>
Re: modifications to ALS.scala
Posted by harini <ph...@gmail.com>.
I am new to development with spark, how do I do that? Can I write up a custom
implementation under package org.apache.spark.ml.recommendation, and specify
"spark-mllib" along with others as a library dependency?
--
View this message in context: http://apache-spark-developers-list.1001551.n3.nabble.com/modifications-to-ALS-scala-tp20167p20169.html
Sent from the Apache Spark Developers List mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe e-mail: dev-unsubscribe@spark.apache.org
Re: modifications to ALS.scala
Posted by Georg Heiler <ge...@gmail.com>.
What about putting a custom als implementation into sparks name space?
harini <ph...@gmail.com> schrieb am Do. 8. Dez. 2016 um 00:01:
> Hi all, I am trying to implement ALS with a slightly modified objective
> function, which will require minor changes to fit -> train ->
> computeFactors within ALS.scala
> <https://github.com/apache/spark/blob/v1.6.2/mllib/src/main/scala/org/apache/spark/ml/recommendation/ALS.scala>
> - Is there a way to do this without having to build spark in its entirety?
> ------------------------------
> View this message in context: modifications to ALS.scala
> <http://apache-spark-developers-list.1001551.n3.nabble.com/modifications-to-ALS-scala-tp20167.html>
> Sent from the Apache Spark Developers List mailing list archive
> <http://apache-spark-developers-list.1001551.n3.nabble.com/> at
> Nabble.com.
>