nucthec-lyc

thanks for your wonderful work! the batch inference now is not supported again? readme shows a n_max_batch parm. but it is not in class_detector.h of this version. please cheching ,thanks...

hello, when i prepare training with a generated pretrained model, and follow darknet yolov3 configure. the total loss is large big because of class_loss is very high ,even to 340+....