This project is a machine learning web application that uses Ensemble Techniques to train multiple models and select the best ones for prediction. The application is developed using Flask, and the trained model is deployed on Render.
Click here to access the deployed application
The goal of this project is to predict employee status (Still Employed or Resigned) based on various factors, such as:
- Age
- Length of Service
- Salary
- Gender
- Marital Status
To achieve high prediction accuracy, the project employs Ensemble Learning techniques, specifically:
- Voting Classifier (Hard Voting)
- Stacking Classifier with a meta-model (Random Forest)
The best-performing models are selected based on accuracy and then used in the ensemble model.
- Python (Machine Learning Model Training)
- Flask (Backend Web Framework)
- Scikit-learn (ML Models & Preprocessing)
- Pandas (Data Handling)
- Joblib (Model Persistence)
- HTML & Bootstrap (Frontend UI)
- Render (Cloud Deployment)
📁 project-root/
│── 📁 model/ # Contains trained models
│ │── Train_model.py # Train multiple models and save the best one
│ │── Load_Train_Best_Model.py # Load and use the best model
│ │── Test_Model.py # Test the trained model
│── 📁 templates/ # HTML templates for Flask
│── 📁 static/ # CSS & assets
│── app.py # Main Flask application
│── data.csv # Dataset used for training
│── requirements.txt # Dependencies
│── README.md # Project documentation
git clone https://github.com/your-repo-url.git
cd project-root
pip install -r requirements.txt
Run the script to train models and save the best one:
python model/Train_model.py
python app.py
The application will be available at http://127.0.0.1:5000/
- Improve model performance using feature engineering
- Implement more advanced ensemble techniques
- Enhance UI design for better user experience
- Deploy on additional cloud platforms