go to Data and download the file named : " cirtaChallenge.ipynb ", good luck.

I've just made this: https://colab.research.google.com/drive/1MrtJE5JhAyy6eZ4JXuEK9Ce3bF6CUhLb

Doesn't do very well, but goes through creating a submission based on predictions (in other words, it covers the whole pipeline).

Scores ~1.6

Feel free to try your own ways of dealing with the data imbalance.

Good luck :)

Thanks