Hi! I used the credit card fraud dataset from Kaggle. I just want to fix the Class column (where 0 = no fraud and 1 = fraud)
It's highly imbalanced
I tried this method
Which got me nowhere
What am i doing wrong?
Did you specify correctly the parameters associated with the size of each class in your sample? (Check the figure below)
Hi @yogipajarillo,
For me, it works with the following settings (chosen arbitrarily) :
pmpmmp
and then the statistics are :
Regards,
Lionel
thank you @earmijo and @lionelderkrikor! i could swear i specified the Sample Size Per Class the first time around, but in any case, i tried again now and the process worked. if you guys had not replied, i would have abandoned the problem and moved on to looking for a different method. thank you again for your time!
