# Competition Timeline
End Date : May 11, 2021 - 11:59 PM EST

## Rules
- You should sign up using either @stonybrook or @cs.stonybrook ID. We will ignore any submissions from other email addresses. You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.
- The public leaderboard will reflect the scores on 30% of the test data. The private leaderboard will evaluate your results on the entire test data and will be made public once the competition gets over.
- You may submit a maximum of 3 entries per day.
- You will have to submit the code (along with the hyperparameters) that you used to generate your best results on the public leaderboard. We might cross-check your Kaggle results with that of your code. So please do not resort to unfair means.
- You may select up to 2 final submissions for judging.
- Only use permitted NLP/ML libraries (see main assignment page; ask if unsure).
- Don't cheat! This is a class assignment: Copying chunks of code or problem-solving answers from other students, online or other resources is prohibited. You are responsible for both (1) not copying others' work, and (2) making sure your work is not accessible to others. Problem-solving solutions are expected to be original using concepts discussed in the book, class, or supplemental materials but not using any direct code or answers (besides permitted libraries). Please see the syllabus for additional policies.
- Experiment! Apply yourself! Use your own ideas or those from class that aren't required for this.
- Have fun!

## Tips (check regularly, we might add a few more!)
Try to generalize your models and avoid overfitting to the training data or even the public leaderboard. With-hold the trial data except for setting hyperparameters or you may overfit. If your models are overfit, expect a significant difference between your results on public and private leaderboard.