I am trying to train a face detection model for my graduate project. I am having two kinds of problems. One is that during training, at one particular stage when the FA(false alarm) goes to 0 , the training hangs(goes in a loop i suppose). How do i prevent this? ( I am giving 500 pos samples and roughly same negative samples). My second problem is, when i train some model with lesser stages(~5) and run them in the detection program, i get an error
Opencv error : GPU Api call NCV assertion failed: NCV stat=28, file cascadeclassifier.cpp line=168.