Hi,I would like to ask a question regarding the classification model.When I use a binary classification model in TMVA package to predict a new testing data.I could get the result that whether it is a signal data or background data. I wonder,when the testing data is predicted as a background data. whether I could simultaneously output the probability that the data is background.Thanks
I’m afraid I don’t quite understand the question (perhaps @moneta does). Can you expand a bit or post illustrative code sample?
I don’t also understand the question. A binary classification problem gives you a score result that you can interpret as a probability that the given event is signal or background.
If your data has the label, so you know its truth class ID, you can then compute signal and background efficiencies (i.e compute what is normally called sensitivity and specificity)