Anant8958/Daimond_Price_predictor

💎 Diamond Price Prediction & Valuation System

An end-to-end Machine Learning web application designed to predict diamond retail prices with high accuracy using a Random Forest Regressor trained on the classic 53,940+ diamonds dataset. The application features an interactive Streamlit user interface allowing users to adjust the 4 C's and diamond proportions in real time to receive immediate valuations and confidence estimates.

---

📌 Project Overview

Diamond pricing is traditionally complex and heavily dependent on specific qualitative and quantitative grading metrics known as the 4 C's (Carat, Cut, Color, and Clarity) alongside geometric table and depth proportions.

This project encompasses:

• Exploratory Data Analysis (EDA): Feature correlations, distribution analysis, and scatter plots comparing diamond dimensions against price.
• Machine Learning Pipeline: Data transformation using ColumnTransformer with OneHotEncoder and modeling with RandomForestRegressor.
• Interactive Web App: A responsive, modern Streamlit UI with cached model deserialization (@st.cache_resource) for instant predictions and zero lag.

---

🚀 Key Features

• Real-Time Price Valuation: Instant price estimates calculated dynamically as you adjust sliders and inputs.
• The 4 C's Controls:
  • Carat: Precise weight control from 0.20 to 5.00 carats.
  • Cut Quality: Fair, Good, Very Good, Premium, Ideal.
  • Color Grade: From D (completely colorless) to J (faint color).
  • Clarity Grade: From I1 (included) to IF (internally flawless).
• Proportions & Dimensional Measurements: Adjustable Total Depth %, Table %, Length ($x$), Width ($y$), and Depth ($z$) in millimeters.
• Valuation Presets: Quick one-click setups for popular diamond types:
  • 1.00ct Engagement Ring (Ideal)
  • 0.50ct Budget Diamond
  • 2.00ct Luxury Solitaire
• Confidence Range: Displays an estimated 95% price range based on the model's Mean Absolute Error ($MAE \approx $270$).

---

📊 Model Architecture & Performance

The model was trained on the standard Seaborn Diamonds dataset (80/20 train/test split, random state 42).

Metric	Score / Value
Algorithm	Random Forest Regressor (n_estimators=100)
$R^2$ Score	~0.981 (98.1% of variance explained)
Mean Absolute Error (MAE)	~$270
Pre-processing	One-Hot Encoding (drop='first', handle_unknown='ignore') for categorical variables, passthrough for continuous numericals

---

📁 Repository Structure

Diamond Price Prediction/
├── app.py                 # Streamlit web application
├── diamond_model.pkl      # Pre-trained Random Forest Regressor model (compressed ~62 MB, GitHub-friendly)
├── encoder.pkl            # Pre-fitted ColumnTransformer preprocessor
├── model.ipynb            # Jupyter Notebook with EDA, training, and evaluation
├── requirements.txt       # Project dependencies
└── README.md              # Project documentation

---

🛠️ Installation & Setup

1. Clone or Open the Repository

cd "Diamond Price Prediction"

2. Set Up a Virtual Environment (Optional, Recommended)

# Windows
python -m venv venv
venv\Scripts\activate

# macOS / Linux
python3 -m venv venv
source venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt

4. Run the Streamlit Application

streamlit run app.py

The app will open automatically in your browser at: 👉 http://localhost:8501

---

📖 How to Use the App

1. Select a Preset or Custom Setup: Use the sidebar dropdown to choose between quick presets or customize all parameters yourself.
2. Configure the 4 C's:
   • Adjust the Carat Weight slider.
   • Select the Cut, Color, and Clarity grades matching your diamond's certificate (e.g., GIA or IGI).
3. Fine-tune Dimensions: Adjust Depth %, Table %, and physical measurements ($x$, $y$, $z$) in millimeters.
4. View Valuation: The estimated retail market price and 95% confidence range will update instantly in the valuation card on the right.

---

📦 Requirements

• Python 3.9+
• streamlit
• scikit-learn
• pandas
• numpy
• seaborn
• matplotlib

---

📄 License

This project is open-source and available under the MIT License. # Diamond-Price-Predator