Dear orgenizers, thanks for what seems a nice competition.
I’ve arrived late, so I hope my contribution would be meaningful.
I am a bioinformatician, and as such, I have relevant knowledge that would help me (hopefully) improve my models. It may be numerical or sequnial features that bear biological meaning. I’d like to ask if that counts as ‘externa data’, since I need to type in some numbers or letters. However, that’s an “accumulated” set of infomation, learnt at the first class of biology101 or acquired during professional experience. I do not use this information for training the model, and it’s not the same case as plugging in additional databases which may contain the private test set. I can generate proxies that resemble these features that are self-contained; it’s likely to harm the model and implemeting it would consume time.
Please let me know what is you opinion on this matter.