Problem with compare Rocs

Catherine
New Altair Community Member
Hello, I'm having trouble seeing the graphical results of compare ROCS in a weather dataset. I'm trying for several days but I can't figure what the problem is if I exclude the problematic attributes it has problems the naive bayes if I include Rain today for example it has problem weight by correlation and SVM but the atrributes exist in the dataset. I removed missing attributes from select attributes but I still have a problem could you please help me?
Tagged:
0
Best Answers
-
Hello Catherine,
Apologies because noone has chimed in. Do you still have those issues? If you can share your RapidMiner file to take a closer look it would be awesome.
All the best, Rod.
1 -
Hi Catherine,
In RapidMiner Studio, with the process open, go to File > Export Process. You can either attach the result as a file or paste it (it's just an XML file, you should be able to open it with your favourite editor).
This way I can download it and see what you're doing.
If you could provide us with data too, it would be nice (however, it's not always the case, therefore if you could at least explain what data you're using, we can experiment with a few values)
All the best,
Rod.0
Answers
-
Hello Catherine,
Apologies because noone has chimed in. Do you still have those issues? If you can share your RapidMiner file to take a closer look it would be awesome.
All the best, Rod.
1 -
Hello and thank you for the help. I've checked normalized weights in weight by correlation and I've removed preprunnning from decision tree and it worked out but I don't know if it is correct..how can I share it?
0 -
Hi Catherine,
In RapidMiner Studio, with the process open, go to File > Export Process. You can either attach the result as a file or paste it (it's just an XML file, you should be able to open it with your favourite editor).
This way I can download it and see what you're doing.
If you could provide us with data too, it would be nice (however, it's not always the case, therefore if you could at least explain what data you're using, we can experiment with a few values)
All the best,
Rod.0 -
I'm using trial version of rapid miner and because I don't want to lose the process (it's for a university study) is it possible to share it in another way?I was thinking if it has to do with the postponing of the update..I didn't update to the latestThank you again0