Here is the updated rules document, explicitly tailored for a **College-Level / University Kaggle Competition**. I have adjusted the eligibility criteria, emphasized academic integrity, and tied the "winner's obligations" to course grading and project reports rather than corporate licensing.

You can copy and paste this directly into your Kaggle InClass or Community competition settings.

---

## ENTRY IN THIS COMPETITION CONSTITUTES YOUR ACCEPTANCE OF THESE OFFICIAL COMPETITION RULES.

*This Competition is an academic, skills-based competition designed to promote practical data science application among university students. You must register via the Competition Website to enter. To enter the Competition, you must agree to these Official Competition Rules, which incorporate your university's Academic Integrity policies. Please read these Rules carefully before entry. Your competition Submissions must conform to the requirements stated on the Competition Website and will be scored based on the evaluation metric described. Leaderboard rankings may be used as a component of your final course grade or project evaluation.*

**You cannot sign up to Kaggle from multiple accounts. Using multiple accounts to bypass daily submission limits is a violation of academic integrity and will result in disqualification.**

---

### 1. COMPETITION-SPECIFIC TERMS

* **1. COMPETITION TITLE:** E-Commerce Retention ROI Challenge
* **2. COMPETITION WEBSITE:** [https://www.kaggle.com/competitions/ecommerce-retention-roi-challenge](https://www.google.com/search?q=https://www.kaggle.com/competitions/ecommerce-retention-roi-challenge)
* **3. DATA ACCESS AND USE:** Academic and Educational Use Only.

---

### 2. COMPETITION-SPECIFIC RULES

In addition to the standard Kaggle Community rules, you understand and agree to the following academic rules:

#### 1. TEAM LIMITS

a. The maximum Team size is 1 student.

#### 2. SUBMISSION LIMITS

a. You may submit a maximum of **five (5)** Submissions per day.
b. You may select up to **two (2)** Final Submissions for final leaderboard grading.

#### 3. COMPETITION TIMELINE

a. Competition Timeline dates (including Entry Deadline, Final Submission Deadline, and Team Merger Deadline) are reflected on the competition’s Overview > Timeline page. **No late submissions will be accepted for leaderboard scoring.**

#### 4. COMPETITION DATA

a. **Data Access and Use.** You may access and use the Competition Data strictly for educational purposes and for participating in this specific class project.
b. **Data Security.** You agree not to distribute, publish, or share the Competition Data outside of the context of this class/competition.

#### 5. ACADEMIC INTEGRITY & CODE SHARING

a. **Original Work:** All submitted models, code, and predictions must be the original work of your team.
b. **Private Sharing Prohibited:** Privately sharing code, data, or predictions with students outside your team is a violation of the university's Academic Honor Code.
c. **Public Sharing:** You *may* share ideas, generic code snippets, and troubleshooting advice publicly on the Kaggle Competition Discussion forum.
d. **Plagiarism:** Copying kernels or notebooks from past Kaggle competitions without citation, or passing off auto-generated code as your sole original work without understanding it, will result in academic penalties.

#### 6. EXTERNAL DATA AND PRE-TRAINED MODELS

a. You may use external, publicly available datasets or pre-trained models (e.g., standard PyTorch/TensorFlow weights) to develop your Submissions, provided they are free, open-source, and posted on the competition discussion board for all other students to access.
b. Paid APIs, proprietary data, or commercial auto-ML tools that require a paid subscription are **prohibited** to ensure a level playing field for all students.

#### 7. GENERATIVE AI & LLM USE

a. You may use Large Language Models (e.g., ChatGPT, Gemini) to assist with writing boilerplate code, debugging, or understanding concepts.
b. However, you must thoroughly understand any code you submit. Your professor/instructors reserve the right to question you on your pipeline; inability to explain your model's architecture or loss function will negatively impact your project grade.

#### 8. ELIGIBILITY

a. This competition is strictly limited to currently enrolled students in at Nirma University.

#### 9. FINAL DELIVERABLES & GRADING OBLIGATIONS

a. To receive full credit for this competition, your team must fulfill the following obligations by the final deadline:

1. **Source Code:** Submit your fully reproducible Jupyter Notebook or Python scripts via the university's learning management system (e.g., Canvas, Blackboard) or as a public Kaggle Notebook post-deadline.
2. **Reproducibility:** Your code must clearly show how you preprocessed the data, defined the Profit-Weighted Log Loss (PWLL) function, trained the model, and generated the final `submission.csv`.

---