# Other Solutions

It is very impressive to read about the top 3 winning solutions. They have pulled away from the rest of the pack by a wide gap.
Some other solutions may still have complimentary components that in combination could offer additional improvement.
Our team came #5.  As others we used CNN. One thing that we found helpful was to vary the aspect ratio of the image. We used both resize preserving and non-preserving aspect ratio, and combined in ensemble. Each alone gave similar scores, but in combination improved by 0.1-0.15.
Others also mentioned using one or the other method. I think this helps for planktons because some of them are very long and thin,  so when resizing with correct AR, not much detail left on the narrow side.
Another method, we used the hierarchical structure to define 4 levels of higher hierarchy labels,.so that the highest level has the fewer number of labels. We trained same network on the separate label sets. We then combined them using the conditioned probability and enforcing some of lower level labels to sum to that of higher ones. An ensemble of such models improved our scores for a given network structure by about 0.1-0.15.
It will be interesting to hear from teams that got improvements from other ideas.
In particular we noticed that in our solution, it did not do as well for classes "unknown" and various "others".  It makes sense, since these are not normal classes but collections of exceptions. Did anyone try a method focused specifically on this aspect, that helped? I was thinking of trying some kind of novelty detection methods, but did not get to that.