Closed laresbernardo closed 3 years ago
I have a hunch that the current target labels might be 1/2 or something like that, and that my function does not detect those as binary classes correctly. Can you share this Titanic dataset of your? Or can I download your exact version somewhere so I can reproduce this error?
I ran into the same error developing today on a different dataset, and I think the bug is solved in the latest commit to the development branch. Could you check whether it now works as planned on your Titanic dataset?
Hi @paulvanderlaken This is a reproducible example.
library(lares) #remotes::install_github("laresbernardo/lares")
data(dft)
ppsr::visualize_predictors(dft, "Survived")
In this case, Survived
is a logical feature and I guess this should be enough to detect binaries, no matter their type.
Hi Lares, thanks for notifying me of this error. I hadn't yet implemented a test for these cases. Pretty basic scenario so should definitely not result in an error! Issue is now solved in latest developmental branch v0.0.0.921 so I will close this issue. Feel free to open again if you run into any errors!
Great package, thanks for sharing @paulvanderlaken I had my eye over this Predictive Power Score methodology for a while now and glad you did this! I am trying to use the package as a plug-and-play function using Titanic dataset but am getting the following error which I think you can avoid by converting character classes into factors in the backend when detected. Is that correct?
I guess you've got the idea as you do something similar here.