# Competition Rules

The goal of this competition is to evaluate your ability to design and implement a machine learning solution under realistic constraints. By participating, you agree to follow the rules below.

## 1. Eligibility

Participation is limited to enrolled students and authorized course staff. If the competition is publicly visible, only approved participants should submit solutions.

## 2. Team Policy

Unless explicitly stated otherwise by the instructor, submissions must be individual work.
If teams are allowed, each participant may join only one team and team membership must remain unchanged after the competition begins.

## 3. Permitted Resources

You may:

* Use standard machine learning libraries and publicly available pretrained models.
* Use publicly available educational resources to learn methods.
* Use the provided training and validation data for experimentation.
* Start from the official baseline notebook.

## 4. Prohibited Actions

You must not:

* Attempt to identify or reverse engineer the origin of the dataset.
* Use external labels or datasets containing overlapping samples.
* Access or attempt to infer hidden test labels.
* Share submission files, predictions, or private solutions with other participants.
* Exploit the leaderboard or submission system to extract hidden information.

## 5. Submission Limits

Participants must respect the submission limits displayed on the competition page. Excessive submissions intended to probe the private evaluation split may be treated as misuse.

## 6. Reproducibility Requirement

Top submissions may be required to provide:

* A short written explanation of the approach.
* A reproducible training and inference pipeline.
* A Kaggle notebook capable of regenerating the final submission.

Failure to reproduce results may result in score invalidation.

## 7. Academic Integrity

This competition follows the university’s academic integrity policy. Plagiarism, unauthorized collaboration, solution sharing, or any attempt to bypass the intended learning objectives may result in disqualification and academic penalties.

## 8. Discussion Policy

Participants may discuss high-level ideas in the competition discussion forum.
Do not share complete solutions, trained model weights, or full submission files.

## 9. Final Ranking

Final rankings are determined by the private leaderboard score.
Organizers reserve the right to review top submissions for compliance with these rules.

## 10. Enforcement

Violations may result in submission removal, disqualification, or additional academic action.
If you are unsure whether something is allowed, contact the course staff before proceeding.

---