Use Monte Carlo data set as input to Predictive model
For R&D purpose, we have created a predictive model with some sample telecom customer data (input - CSV). Now, we are trying to change the input to Monte Carlo data set.
I've created Monte Carlo data set with the required columns (10K records). When this data set is referred in Data selection step, I see only two rows populated out of which the first row is the data set fields itself.
Can you please suggest how can we get this work. PFA snapshots of data set and data selection page.
Now, I have created DDS data set, set it as destination in data flow and ran it. Now when I select DDS data store as input to Predictive model, columns are properly displayed by only one record is shown in the data selection screen.
PFA snapshots. Do I need to select any of the save options? Am I doing anything wrong.
Please verify that you do get records in the DDS set. If there's just one you may have used an incorrect key perhaps, so that all records have the same key value?
The data flow should be trivial
With the MC source for example like this
Make sure the destination dataset is properly keyed. For example:
Now you can just use that in PAD:
NOTE: Screen shots here are from the upcoming 7.4 version. I'm getting an error as well when trying to use the MC dataset directly (we'll try to address this for this version). But with the DDS set it works for me.
PS my Pega ID is "perdo" but my first name is Otto
I guess I found the problem. As you said issue is with the Key. I have used combined key for the data class which are Area code and Mobile number. But I am not able to generate values for these two fields from MC data set as per our requirement. I have posted the same on a different thread "https://pdn.pega.com/community/product-support/question/queries-regardi…". But I haven't received any response on it yet. Can you please check if you can help me out with the queries post on the other thread.