It may take some work to do online updates with an
MatrixFactorizationModel because you need to update some rows of the
user/item factors. You may be interested in spark-indexedrdd
(http://spark-packages.org/package/amplab/spark-indexedrdd).
We support save/load in Scala/Java. We are going to add Python support
for model import/export soon. As a hack, try the following in Python:
model = ALS.train(...)
model.call("save", sc._jsc.sc(), "/tmp/als") # save model to somewhere
newModel =
MatrixFactorizationModel(sc._jvm.org.apache.spark.mllib.recommendation.MatrixFactorizationModel.load(sc._jsc.sc(),
"/tmp/als"))
Best,
Xiangrui
On Thu, Feb 26, 2015 at 10:42 AM, anishm <anish.mashan...@gmail.com> wrote:
> I am a beginner to the world of Machine Learning and the usage of Apache
> Spark.
> I have followed the tutorial at
> https://databricks-training.s3.amazonaws.com/movie-recommendation-with-mllib.html#augmenting-matrix-factors
> <https://databricks-training.s3.amazonaws.com/movie-recommendation-with-mllib.html#augmenting-matrix-factors>
> , and was succesfully able to develop the application. Now, as it is
> required that today's web application need to be powered by real time
> recommendations. I would like my model to be ready for new data that keeps
> coming on the server.
> The site has quoted:
> *
> A better way to get the recommendations for you is training a matrix
> factorization model first and then augmenting the model using your ratings.*
>
> How do I do that? I am using Python to develop my application. Also, please
> tell me how do I persist the model to use it again, or an idea how do I
> interface this with a web service.
>
> Thanking you,
> Anish Mashankar
> A Data Science Enthusiast
>
>
>
> --
> View this message in context:
> http://apache-spark-user-list.1001560.n3.nabble.com/How-to-augment-data-to-existing-MatrixFactorizationModel-tp21831.html
> Sent from the Apache Spark User List mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscr...@spark.apache.org
> For additional commands, e-mail: user-h...@spark.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscr...@spark.apache.org
For additional commands, e-mail: user-h...@spark.apache.org
