Let me chuck my 5 cents into the discussion in regards the future intentions.
Basing on the foundation (distributed math implementation), perfectly described by Nikita, Ignite community is going to build widely and coommonly used algorithms like classifications, regressions, k-means clustering, decision trees, etc. The first algorithms are to be available in the 2.1. If to jump even further to the future, Ignite will have Python and R libs in its ML stack. That's a must for scientists and researchers. As for the use cases - real-time and near-real time calculations basing on trained models harnessing the whole cluster's resources. For instance, here you can find some of the business use cases: https://www.google.com/amp/s/www.forbes.com/sites/bernardmarr/2016/09/30/what-are-the-top-10-use-cases-for-machine-learning-and-ai/amp/ Denis On Saturday, May 6, 2017, Nikita Ivanov <[email protected]> wrote: > Let me try to answer here. > > (1) Apache Ignite 2.0 is coming with the early (beta) version of ML Grid. > ML Grid (beta) will have basic core math (vectors and matrices) for local > and distributed, dense and sparse processing on Ignite cluster. It will > also have all the basic surrounding block. The design is heavily borrowed > from Apache Mahout but adopted to Apache Ignite in-memory platform. > > (2) I know many users of the project use Ignite as-is for speeding up > matrix and vector operations even before ML Grid. You can use Compute and > Data Grids together to develop highly efficient co-located processing of > sparse data sets. You can your training data, for example, in Data Grid and > use custom Compute Grid tasks to process it. Obviously, with introduction > of ML Grid that should be a lot simpler now... > > Hope it give you couple of pointers on where to start. > > Best, > -- > Nikita Ivanov > > > On Sat, May 6, 2017 at 2:04 AM, ChickyDutt <[email protected] > <javascript:_e(%7B%7D,'cvml','[email protected]');>> wrote: > >> Hi team, >> >> I want to know how and where could Apache ignite fit in the field of >> predictive analytics and machine learning. >> >> Do you have any use cases or reference that outline the interaction >> between the two aforementioned systems. >> >> Regards. >> >> ------------------------------ >> View this message in context: Machine learning and Apache Ignite >> <http://apache-ignite-users.70518.x6.nabble.com/Machine-learning-and-Apache-Ignite-tp12470.html> >> Sent from the Apache Ignite Users mailing list archive >> <http://apache-ignite-users.70518.x6.nabble.com/> at Nabble.com. >> > >
