id summary reporter owner description type status priority milestone component version resolution keywords cc 2755 For gradient boosted trees with LogisticRegressionLoss the threshold for the discriminant function should be set to zero. gkronber mkommend Currently, the 'accuracy maximizing' threshold calculator is used. However, this is rather slow for many data points and additionally it introduces a potential problem of overfitting. The pragmatic way is to set the threshold to zero. enhancement closed medium HeuristicLab 3.3.15 Algorithms.DataAnalysis 3.3.14 done