Incremental classification of SVM or any classifier in SPARK

Question

Is there any way to perform incremental classification of any classifier in Spark / MLlib ? What I want is to retrain existing model with new data set. New data set can come at any time and i would like to add it in already trained classifier.

Answer 1

Yes, Spark provides StreamingLogisticRegressionWithSGD in spark.mllib package with which you can train or predict a logistic regression model on streaming data. For further details please refer spark documentation.