Closed jemus42 closed 2 years ago
Nice and clean PR, thanks! 👍
Thanks again @jemus42!
I've noticed we're missing the paramtest for this learner - do you want to add it yourself? Maybe @mllg can assist in the process?
Oh no, I thought I had everything /o\ I'll take a look and open a new PR tomorrow.
As noted on Mattermost, it seemed odd this learner was omitted until now.
This PR adds
LearnerRegrNnet
as mostly a verbatim copy ofLearnerClassifNnet
, with the notable exception that the parameterlinout = TRUE
is forced during training to ensure a linear output suitable for regression tasks.