you can filter the data you need by using PEventStore API or modifying the DataSource.scala. Then train different engine instance with different configuration.
On Wed, Sep 7, 2016 at 4:57 PM, Donald Szeto <[email protected]> wrote: > Hi, > > If you are using the stock classification engine template, your easiest > way out would be to use multiple engine servers. If that is not desired, > you could modify the engine template to handle different data sets > independently. > > Regards, > Donald > > On Tue, Sep 6, 2016 at 5:49 AM, Dipen Patel <[email protected]> wrote: > >> >> >> Hello, >> >> I am interested in using predictionio's classification engine. Would the >> classification be able to accurately classify if it had data consisting of >> multiple independent sets of data. For instance, if a training data set >> accepted the id corresponding to a movie director, the amount of money >> spent filming the movie, and the views the movie received, could one >> classification engine be able to predict the amount of views a movie would >> be likely to receive if the training data had included data of different >> directors? My concern is that not having a separate classification engine >> instance for each director will cause the single classification engine to >> mistakenly use data from other directors. >> >> Thanks. >> >> >
