You don't have to implement algorithms from scratch, but you can use Compute Grid to parallelize existing algorithms using existing libraries. Compute Grid API is fairly simple, so if the algorithm itself allows to run it in parallel, this should not be hard to implement.
-Val -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Ignite-support-machine-learning-librarys-tp4305p4427.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.
