You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The Multiclass Classification catalog has non calibrated trainers (SdcaNonCalibratedMulticlassTrainer) along with many calibrated trainers.
Currently there is only one Evaluate method, which assumes that the model is calibrated, and therefore produces LogLoss metrics. These metrics don't makes sense for the non calibrated trainers.
We should therefore add an evaluation method named EvaluateNonCalibrated for the non calibrated trainers.
The text was updated successfully, but these errors were encountered:
The Multiclass Classification catalog has non calibrated trainers (
SdcaNonCalibratedMulticlassTrainer
) along with many calibrated trainers.Currently there is only one
Evaluate
method, which assumes that the model is calibrated, and therefore producesLogLoss
metrics. These metrics don't makes sense for the non calibrated trainers.We should therefore add an evaluation method named
EvaluateNonCalibrated
for the non calibrated trainers.The text was updated successfully, but these errors were encountered: