05-28-2017 03:44 AM
I have a question about "model scoring" task. I have created and registered a model in Enterprise Miner, now I am scoring this model from Enterprise Guide using "model scoring" task.
In the output variables of the model I have three types of var:
P_targetvar Q_targetvar U_targetvar V_targetvar
I think that P_targetvar is the predicted value for target variable of the model, but I would like to know the
meaning of Q_, U_ and V_ vars.
Another question I would like to know if I can definde and use a Enterprise Miner model with parameters. I mean not only
the vars of the dataset, this parameters would be about the type of model (selection criteria, type of decision tree...)
Last question I would like to know why enterprise miner make the partition in three datasets: train, validation and test. I know tht train is for training the modelo and validation to valid model predictions, but I don't know the use of test.
Thanks in advance