I am developing a siamese network for face recognition . I am thinking of using the InceptionV3 module in keras as the CNN architecture. But when I try to concatenate the la
