#  Relationship Status Prediction Challenge  
**GDGC NIT Jalandhar - AI/ML Inductions Kaggle Competition**

---

##  Problem Statement  
Predict the **relationship probability (0–100 scale)** of students at NIT Jalandhar based on various personal, academic, and social features.  
This is a **regression problem** where the goal is to predict a continuous value representing the likelihood of a student being in a relationship.

---

##  Dataset Description  

###  Files Provided

- **`train.csv`** – Training dataset with features and target variable  
  - 5,000 samples  
  - 33 encoded features (`F1`–`F33`) + `ID` + `relationship_probability` (target)

- **`test.csv`** – Test dataset with features only  
  - 2,000 samples  
  - 33 encoded features (`F1`–`F33`) + `ID`

- **`feature_lookup.csv`** – Feature name mapping and metadata  
  - Maps encoded features (`F1`, `F2`, etc.) to actual feature names  
  - Shows feature type (`numeric` / `categorical`)

- **`sample_submission.csv`** – Sample submission format  
  - Required structure: `ID`, `relationship_probability`

---

##  Important Notes

- **Feature Names are Encoded**  
  All feature names in `train.csv` and `test.csv` are encoded as `F1`, `F2`, `F3`, etc.  
  Use `feature_lookup.csv` to understand what each feature represents.

- **Distribution Shift**  
  The training and test sets have **different target distributions**.  
  Simple train-test split validation may not reflect actual performance!

- **Irrelevant Features**  
  Not all features are equally important.  
  Some features are **noise** and won't help your model.

---

Happy modeling!