Are we allowed to augment data in Logo Classification problem. And how are we supposed to submit. Are we supposed to push our code here.
Also what Test set accuracy are you expecting for the Flickr Logo Dataset, because it's not very much even by the research papers which have cited this dataset