Open kavir1698 opened 7 years ago
Thank you for the report, I seem to have neglected porting that code over. Would you like to make a PR? It should be a straight-forward modification of scorer.jl. You'd have to translate this code from Python. I'm happy to help out with the details if you're interested.
If not, this may work as a work-around until that function is implemented:
using ScikitLearn
@sk_import metrics: r2_score
r2_scorer = ScikitLearn.Skcore.make_scorer(r2_score)
cross_val_score(lr, X, y, scoring=r2_scorer)
Yes, I will give it a shot.
The following code gives the error below:
Any other scoring method will give the same error, except the "mean_squared_error", of course. The same code works in Python.