🧠 Machine Learning Pipeline

The pipeline includes a rent prediction model built with scikit-learn.

---

⚙️ ml_configs.py

• Centralized ML configuration (model paths, hyperparameters).

---

📥 data_loader.py

• Fetches structured property + floor plan data from PostgreSQL.
• Cleans and prepares features:
• Bedrooms, bathrooms, sqft, reviews, year built, etc.

---

✨ preprocessor.py

• Defines transformation pipeline:
• StandardScaler for numeric features
• OneHotEncoder for categorical features
• Built using ColumnTransformer.

---

🏋️ trainer.py

• Splits data into train/test.
• Trains Linear Regression model.
• Evaluates using MSE and R².
• Saves pipeline (preprocessor + model) into .pkl.

---

🎶 main.py

• Orchestrates ML pipeline:
• Load → preprocess → train → evaluate → save model
• Provides error handling.
• Designed for scheduled retraining.

---

🔮 Prediction Flow

• Model is loaded at FastAPI startup.
• API endpoint /predict/rent uses the trained pipeline.
• Returns rent price predictions in real-time.