spark-mllib-streaming.adoc

Streaming MLlib

The following Machine Learning algorithms have their streaming variants in MLlib:

• k-means

• Linear Regression

• Logistic Regression

They can train models and predict on streaming data.

Note	The streaming algorithms belong to spark.mllib (the older RDD-based API).

Streaming k-means

org.apache.spark.mllib.clustering.StreamingKMeans

Streaming Linear Regression

org.apache.spark.mllib.regression.StreamingLinearRegressionWithSGD

Streaming Logistic Regression

org.apache.spark.mllib.classification.StreamingLogisticRegressionWithSGD

Sources

• Streaming Machine Learning in Spark- Jeremy Freeman (HHMI Janelia Research Center)