A Tensorflow model for text recognition (CNN + seq2seq with visual attention) available as a Python package and compatible with Google Cloud ML Engine.
If we train on data having simple labels like A-Z and 0-9, what does the model predict for negative image(say blank image), (please forgive me if I sound dumb, just a thought) having no labels associated with it? Also in training, can we include negative class( X = blank image , Y= no label).
If we train on data having simple labels like A-Z and 0-9, what does the model predict for negative image(say blank image), (please forgive me if I sound dumb, just a thought) having no labels associated with it? Also in training, can we include negative class( X = blank image , Y= no label).