Large Scale Machine Learning (Big Data): Difference between revisions

<code>Snap ML</code> features distributed implementations of many estimators. To run in distributed mode, call your python script using <code>mpirun</code> or <code>srun</code>.

[https://spark.apache.org/docs/latest/ml-guide.html Spark ML] is a Machine Learning library built on top of [[Apache_Spark/en|Apache Spark]]. It enables users to scale out many Machine Learning methods to massive amounts of data, over multiple nodes, without worrying about distributing datasets or explicitly writing distributed/parallel code. The library also includes many useful tools for distributed Linear Algebra and Statistics. Please see our tutorial on [[Apache_Spark/en#Usage|submitting Spark jobs]] before trying out the examples on the official [https://spark.apache.org/docs/latest/ml-guide.html Spark ML documentation].