Classification - HELP

So i've got two data sets, a training set and a test set (both csv files). My aim is to predict the class labels of the test set. I've tried doing it but keep getting various errors / don't even know if i'm going about it the right way. The issue I initially faced was that the test data didn't have ALL the columns that the training data does. Can anybody help guide me in the right direction? Can give more info if needed! 

 

Many thanks!

 

Hello rsalam,

To get started the Online Self-Training is a good source. Also there are multiple Youtube-Videos and of course the example server within the Analytics Platform.

Regarding the mentioned error:
Your Learner is trained on the features given in the training set (for example a Learner might generate rules like "If temperature less than 20 heat is on) and then the model is applied by the Predictor on the test set. If the test-set is missing some features some predictors have problems applying the model (How to determine whether heat is on if the temperature is not given?). Therefore you have to determine if you can derive these features from the ones given in your test set, or you might have to abandon them (Column Filter).

Best,
Ferry