General Rules
Team Composition:

Participants can compete only in teams (of 2 ).
Teams must not share code, data, or strategies with other teams.
Submission Limits:

Each team is allowed up to 5 submissions per day.
Only the last submission of the day will count toward the leaderboard.
Competition Timeline:

Submissions will be accepted until the competition deadline.
Late submissions will not be considered under any circumstances.
Ethical Participation:

Participants must not attempt to hack the competition platform or manipulate the leaderboard.
Use of unethical practices, such as falsifying submissions or using prohibited external data, is strictly prohibited.
Dataset Rules
Provided Dataset Only:

Participants must only use the dataset provided as part of the competition.
***External labeled datasets are not allowed.*** and may result in zero Grade
Using external pre-trained models or any models learnt in the course or outside the course  is allowed
No Data Augmentation with Labeled Data:

You may perform data augmentation using techniques like translation or paraphrasing, but additional labeled data cannot be created externally.
Model and Code Rules

One Account per Participant/Team:

Each participant/team can have only one Kaggle account for the competition.
Multiple accounts for a single participant/team will result in disqualification.
Submission Rules
Format:

Submissions must follow the prescribed format (Tweet_ID,Label).
Improperly formatted submissions will be rejected.
Final Evaluation:

The leaderboard will display scores on a public subset of the test set during the competition.
In Kaggle competitions, participants are ranked on two leaderboards: the Public Leaderboard and the Private Leaderboard. During the competition, the Public Leaderboard displays rankings based on a subset of the test data, allowing competitors to gauge their model's performance and adjust accordingly. ***However, final rankings are determined by the Private Leaderboard, which evaluates models on a separate, hidden portion of the test data.*** This mechanism ensures that participants cannot overfit their models to the Public Leaderboard and promotes the development of more generalizable solutions.

Submission Integrity:

Submissions must be generated by the participant/team's model.
Manually crafted predictions or external assistance is not allowed.
Disqualification
Participants may be disqualified for:

Violating competition rules (e.g., using external labeled datasets).
Falsifying submissions or manipulating the leaderboard.
Sharing code, strategies, or models with other teams during the competition.