NicolasC,
There is a lot of great information in the Enterprise Miner help which will likely help you. If you open SAS Enterprise Miner and click on Help --> Contents, you can search on _WARN_ -- the excerpt below is from the Predictive Modeling document which appears first on the list:
/*** BEGIN EXCERPT ***/
A variable named _WARN_ in the scored data set indicates why the model could not be applied. If you have lots of cases with missing inputs, you should either use the Decision Tree node for modeling, or use the Impute node to impute missing values before using the Regression or Neural Network nodes.
/*** END EXCERPT ***/
This variable is added at modeling nodes to the data sets that pass through. It should only be a concern if there is a non-missing value in the field. In many cases, the field might be present yet not have any value/text for any of the observations. If you are encountering observations where _WARN_ is not missing, please let us know what nodes you are using prior to the Bayesian Network along with the values you are seeing.
Hope this helps!
Doug
