1. Eligibility & Team Formation
- Open to students of JBIET and invited institutions (as notified by organizers).
- Each team must consist of 3–4 members.
- All members must be registered via the official Google Form before the deadline.
- A participant can be part of only one team.
2. Problem Statements & Tracks
- Tracks include: NLP, Time Series & Forecasting, Computer Vision, Generative AI.
- Final problem statements / datasets will be provided on the day of the hackathon.
- Teams must build their own models from scratch using the given data (no pre-trained models unless explicitly allowed by organizers).
- External public datasets are allowed only if organizers approve and if they are clearly cited.
3. Development Rules
- Coding must be done during the event time only.
- Teams may use any programming language and ML libraries (Python preferred: NumPy, pandas, scikit-learn, PyTorch, TensorFlow, etc.).
- Use of AI tools (e.g., ChatGPT, Copilot) is allowed for assistance, but final logic must be understood by the team.
- All code should be original to the team and not copied from existing GitHub projects without significant modification and understanding.
4. Evaluation & Judging Criteria
- Model Performance – accuracy / error metrics appropriate to the task.
- Approach & ML Design – feature engineering, model selection, experimentation.
- Code Quality – clarity, reproducibility, structure, and documentation.
- Presentation & Explanation – clarity in explaining decisions, results, and trade-offs.
- Originality – creativity in approach and solution design.
5. Submission Requirements
- Teams must submit:
- Code (Jupyter notebook and/or Python scripts).
- A short readme / report describing approach, model, and results.
- Any configuration or requirements file (if needed to run the code).
- Submissions after the deadline may not be evaluated.
- Each team must be ready with a short demo / presentation (5–7 minutes).
6. Conduct & Fair Use
- Plagiarism (code or ideas) is strictly prohibited and may lead to disqualification.
- Teams must behave professionally and respectfully towards organizers, mentors, judges, and other teams.
- Any attempt to manipulate results, misuse data, or violate rules will lead to immediate disqualification.
- Decisions of the judging panel and organizers will be final and binding.
7. Logistics & Setup
- Participants should bring their own laptops, chargers, and necessary accessories.
- Internet access will be provided as per institute policy; downloading large models / datasets may be restricted.
- Teams are responsible for regularly saving their work and maintaining backups.
8. Queries & Support
For any clarification related to rules, problem statements, or logistics, please contact the Student or Faculty Coordinators listed on the main page.
Note: Organizers reserve the right to update or refine these guidelines before or during the event. Any changes will be communicated to all teams.