# FDP - Data Science Pipeline: End-to-End Workflow

**Topic:** Data Science Pipeline | **Host:** Jodhpur Institute of Engineering & Technology (JIET)

## 📂 Contents

| File | Description |
|------|-------------|
| `fdp-colab-content/student_stress_data.csv` | Synthetic dataset of 300 student records for stress level prediction |
| `fdp-colab-content/FDP_Student_Stress_Prediction.ipynb` | Google Colab notebook — zero setup required |
| `fdp-colab-content/FDP_PreRead_StudentStress.pdf` | Pre-read material for FDP attendees |
| `fdp-colab-content/generate_dataset.py` | Dataset generator script |
| `fdp-colab-content/pre_read.html` | Pre-read material source (HTML) |

## 🎯 Dataset Overview

**300 student records × 8 columns**
- **Target:** `Stress_Level` (0 = Not Stressed, 1 = Stressed)
- **Features:** Department, Attendance_Level, Favorite_Color, Weekly_Study_Hours, Assignment_Submission_Pct, Backlogs_Count

## 🚀 Using the Colab Notebook

1. Open the notebook in [Google Colab](https://colab.research.google.com/)
2. Run all cells — no installation needed
3. Dataset auto-loads from raw GitHub URL

## 📊 Pipeline Steps

1. Data Loading
2. Exploratory Data Analysis
3. One-Hot Encoding (Department, Attendance_Level, Favorite_Color)
4. Random Forest Classification
5. Model Evaluation
6. Feature Importance Visualization

---

*Part of Gyannetra's FDP workshop series*