## EVALUATION 
The leaderboard uses the fraction of samples predicted correctly as the metric for evaluation.
You will first be shortlisted based on your model’s accuracy. We will also evaluate you based on the originality of your code and its interpretability. Any plagiarism would be penalised.