# 🏋️‍♂️ Gradient Boosting Deadlift Predictor

This repository contains the winning model from Assignment #2: Classification, Regression, Clustering & Evaluation.

## 📌 Model Purpose
The model predicts an athlete's **deadlift performance (lbs)** based on physical and strength-related features.

## 🧠 Algorithm
✅ Gradient Boosting Regressor  
Selected as the final model after comparing:

- Linear Regression
- Random Forest
- Gradient Boosting

## 🏆 Performance (Test Set)

- R²: 0.85
- MAE: ~28.6 lbs
- RMSE: ~37.2 lbs

Gradient Boosting achieved the **highest accuracy and lowest error**, so it was chosen as the final model.

## 📁 Files
- `winning_model.pkl` – serialized model ready for loading and inference

## 🔧 Usage
```python
import pickle

with open("winning_model.pkl", "rb") as f:
    model = pickle.load(f)

prediction = model.predict([[weight, height, backsquat, snatch]])