# Private LB 13th Solution - so near yet so far :)

First of all, congratulation to all, and especially to the prize winner and gold medalists.
Secondly, ouch, narrowly missing out on gold.
So here are some essential points of our solutions:
DL framework: Pytorch
Used Architecture: Effnet(timm/geffnet) B0/B2/B3/B4/B5,
[Rexnet](https://github.com/clovaai/rexnet)
1.3/1.5/2.0
Model customization:
[@haqishen](https://www.kaggle.com/haqishen)
and
[@garybios](https://www.kaggle.com/garybios)
did some amazing work in adding convolution head and SE head to our chosen architecture, and dropping the last two blocks from before the final FC output also helps - I will let them expand on these.
Upsampling Cover: We find that upsampling cover image to match amount to Stegos helps to improve both CV and LB
Method of ensemble: we have trained about 80+ models, we have taken the top models with top 6CV in each fold, and did unweighted gmean
We were particularly careful about overfitting the LB so throughout the competition we didn't use it for feedback. But in general, we observed that for the same fold of data, better CV usually (not always) yields better LB
Best Single Model - B5 with upsampling covers, added convolution heads, drop blocks -> Private LB 0.927, Public LB 0.928
Congratulation to my deal teammates
[@garybios](https://www.kaggle.com/garybios)
[@haqishen](https://www.kaggle.com/haqishen)
[@strideradu](https://www.kaggle.com/strideradu)
and
[@yl1202](https://www.kaggle.com/yl1202)
for the effort, I really had a great time in this short and yet intensive competition! :)