🏠 House Price Prediction API

A Machine Learning API for predicting California housing prices, inspired by Chapter 2 of Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Aurélien Géron. This project walks through the full pipeline, from data preprocessing to prediction, exposed as a simple API using FastAPI.

---

📚 Project Overview

This API provides housing price predictions based on longitude, latitude, housing median age, total rooms, total bedrooms, population, households, median income, and ocean proximity. It encapsulates all the steps from the book’s chapter including:

• Data cleaning & transformation
• Custom feature engineering
• Model training using RandomForestRegressor
• Pipeline serialization using joblib
• And finally, exposing predictions via an API

---

🛠 Tech Stack

• Language: Python 3.12
• Framework: FastAPI
• ML Library: Scikit-learn
• Data Handling: Pandas, NumPy
• Visualization: Matplotlib
• Serialization: Joblib

---

⚙️ Installation

# Clone the repository
git clone https://github.com/aliiakbarkhan/house-price-prediction-api.git
cd house-price-prediction-api

# Create virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate  # on Linux/Mac
venv\Scripts\activate     # on Windows

# Install dependencies
pip install -r requirements.txt

# Run the API
uvicorn main:app --reload

🧪 How to Use the API

Once running, the API will be available at:

http://127.0.0.1:8000/docs

📥 Post / Predict

Send a JSON payload like:

{
  "longitude": -122.23,
  "latitude": 37.88,
  "housing_median_age": 41.0,
  "total_rooms": 880.0,
  "total_bedrooms": 129.0,
  "population": 322.0,
  "households": 126.0,
  "median_income": 8.3252,
  "ocean_proximity": "NEAR BAY"
}

📤 Response

{
  "prediction": 452100.0 (which is $4,52,100)
}

📊 Graphs and Visuals

Plot Description	Image
Geographical Plot
Pairplot: Every numerical feature against every other
Histograms for each numerical feature
Jet-colored housing prices by location and population

🧠 Model

• Algorithm Used: Random Forest Regressor.
• Data Source: California Housing Dataset.
• Feature Engineering: Custom transformer CombinedAttributesAdder for feature addition.
• Evaluation Metrics: RMSE (Root Mean Squared Error).

📁 Project Structure

├── datasets                 # Datasets Folder
├── graphs                 # Visual Graphs Folder
├── notebooks                  # Jupyter Notebook Folder
├── main.py                  # FastAPI app file
├── custom_transformers.py  # Building Pipeline file
├── model.pkl               # Trained RandomForest model
├── pipeline.pkl            # Full preprocessing + model pipeline
├── requirements.txt        # Project dependencies
└── README.md               # Project documentation

🧪 Development & Learning Source

This project is inspired by the practical walkthrough in Chapter 2 of Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Aurélien Géron.

👨‍💻 Author

• Ali Akbar Khan
• Email: aliakbarkhana79@gmail.com
• LinkedIn: aliakbar-khan

➕ Additional Features

• This project goes beyond the textbook by converting the model into a real-world FastAPI-based API.

🔒 License

LICENSE.mp4