guidocarvajalo
Hi,I have the following issue. I need to test a model with a case file and I want to calculate some performance metrics from the output probabilities obtained for a target node. I know Bayesia Lab has the Analysis>Network performance>Target to do this, but it only has limited gauging metrics. I also know that from cross-validation option we can get the raw output but in this case I don't want to use cross-validation. Is there any way I can obtain the probabilities?. Many thanks in advance.Regards,Guido
Quote 0 0
Dan
Hi Guido,You just have to use Inference | Batch Inference to get the posterior probability distribution of your target for each observation contained in your dataset.Hope this helps,Dan
Quote 0 0
guidocarvajalo
Thank you very much Dan, it's great to know that B.L has this optionRegards,Guido
Quote 0 0