GDG-DevSphere-Hard

Problem Statement

In the wake of increasing digital and physical footprints across the IIIT Lucknow campus, the administration has noticed several "Phantom" anomalies that standard monitoring systems cannot explain. As a Data Detective, you must build a Tactical Orchestration Layer using the phantom_tools library. Your goal is to chain multiple tiers of campus data—ranging from WiFi access points to financial transactions—to identify suspicious patterns, generate supportive outreach, and visualize campus resource usage.

---

The Investigation: 3 Core Tasks

Task 1: The Fiscal-Academic Outreach

• The Mystery: Identify students who are "silent strugglers"—those facing both low academic engagement and financial constraints.

• Criteria: Find students with Attendance < 75% AND Campus Cash Balance < ₹500.

• Final Output: A GenAI (LLM) Call. Use the filtered student data (Major and CGPA) to prompt an LLM to generate a personalized, empathetic counseling email for the identified students.

Task 2: The Morning Ghost Routine

• The Mystery: Detect "Digital Vampires" in the 2024 Batch who show heavy network activity but are physically absent during morning routines.

• Criteria: Find students with 0 breakfast_swaps AND Daily Bandwidth > 5GB.

• Final Output: A Python List of Tuples. Format: [(student_id, major, daily_bandwidth), ...].

Task 3: Library Congestion Study

• The Mystery: Determine if library overcrowding is caused by academic research or high-bandwidth streaming/gaming.

• Criteria: Filter for students connected to the "LIBRARY_MAIN" Access Point. Group them by Major and calculate the average bandwidth for each group.

• Final Output: A Matplotlib Bar Chart. The chart must display the Average GB consumed per Major to identify which department is "squatting" on the network.

---

Setup and Installation

Fork & Clone

To begin the challenge, you must create your own copy of the project:

1. Fork this repository by clicking the "Fork" button at the top right of this page.

2. Clone your forked repository to your local machine:

   Bash

   git clone https://github.com/YOUR_USERNAME/DevSphere-ML-Hard.git
   cd DevSphere-ML-Hard
   

---

To ensure all scripts run smoothly without conflicting with your global Python installation, please follow these steps to set up a Virtual Environment:

Create the Virtual Environment

Open your terminal in the GDG-DevSphere folder and run:

Bash

# On Windows
python -m venv venv

# On macOS/Linux
python3 -m venv venv

Activate the Environment

Bash

# On Windows (Command Prompt)
venv\Scripts\activate

# On Windows (PowerShell)
.\venv\Scripts\Activate.ps1

# On macOS/Linux
source venv/bin/activate

---

Install Dependencies

Once activated, install the required libraries for the hackathon:

Bash

pip install -r requirements.txt

---

The Ragathon Mentor Bot

If you get stuck on the logic or don't know which API method to call, run the mentor_bot.py script.

What the bot DOES:

• Method Blueprints: It will tell you the exact sequence of methods (e.g., get_all_student_ids() → get_attendance()).

• Logic Guidance: It explains how to chain data from one tier to another.

What the bot DOES NOT do:

• Write Code: It is strictly forbidden from giving you Python snippets, loops, or logic implementation.

• Give Answers: It won't reveal the "Hidden Truths" (the specific suspicious IDs) in the database.

---

Project Structure

• campus_database.json: The core data source for all tasks.

• phantom_tools.py: The official library containing Tiers 0-5 of the campus API.

• task1.py, task2.py, task3.py: Your skeleton files. Fill in the TODO sections to solve the mysteries.

• .env: Store your LLM API keys here (e.g., GROQ_API_KEY=your_key_here).