SIIM-ISIC Melanoma Classification (6th rank approch)
We started this competition after 2 months after the start. For a baseline, we used @cdeotte's Triple Stratified Kfold With Tfrecords for tensorflow and @shonenkov 's Training CV Melanoma Starter for PyTorch. Thanks to these amazing kernels.
We used EfficientNet [B0-B6], Resnest,Resnext, with Sizes 192x192 256x256 384x384 512x512 768x768 384x512[HxW]
- Heavy TTA (X20)
- Cutmix
- Coarse dropout
- SWA(Stochastic Weight Averaging)
- Loss-Label Smoothing, BCE
- Optimizers - AdamW, Adam
- 2018, 2020 and malignant datasets
- 5 checkpoints' prediction averaging(stabalised our model's predictions)
- some models were trained with different height width ratios
- Loss functions-Focal loss, dice loss
- Optimizer- Ranger
- Hair removal/addition
- Pseudo labelling
- 2019 dataset
- Preprocessing techniques from Aptos Competition
- Progressive learning
- Weighted average
- Power Average
- Minmax ensemble(didn't help)
3hr before end of competition we came across rank ensembling and and we did this ensemble and got 0.9697 for our last submission
We new the shakeup was coming, so we tried to select different approaches.
- All 15+ pytorch gpu solution models(with context) - 0.9530 (public LB) 0.9380 (private LB) 0.9541 (CV)
- 15+ pytorch model (with context) and 15+ tf models (without context) - 0.9627 (public LB) 0.9470 (private LB) 0.9618 (CV)
- Blend of public submission with 2nd submission with post proccessing technique - 0.9697 (public LB) 0.9126 (private LB) (overfitted) all the above were also ensembled with the meta only submission.
We wanted to give a shot to public lb overfitted submission but obviosly didn't work out well. I guess we were lucky enough to select the best private lb submission from our arsenel. We found the discussions and public kernels really fruitful and learnt a lot from this competition.