Closed cadrev closed 3 years ago
NGBoost gives you back a full predicted distribution. If you want to get metrics for point predictions, you should first extract the point predictions (e.g. by taking the mean of the predicted distributions for each observation) and then use standard packages (e.g. sklearn.metrics
) to compute the metrics you want.
Hello, I'd like to ask if there is a way to compute both C-Index and Brier Scores for the survival predictions?