Salary Regression Model

📊 Dataset

• Gender: 0 (Female) or 1 (Male)
• Age: Age in years
• PhD: 0 (No) or 1 (Yes)
• Salary: Target variable (in thousands)

🚀 Quick Start

Installation

Usage

1. Exploratory Data Analysis (EDA)
   python eda.py
   Generates visualization plots:
   • salary_distribution.png
   • correlation_matrix.png
   • salary_by_gender.png
   • salary_by_phd.png
   • pairplot.png
2. Train Models
   python train_model.py
   Trains and compares:
   • Linear Regression
   • Polynomial Regression (Degree 2)
   Saves the best model as best_model.pkl
3. Make Predictions
   python inference.py
   Interactive script for making salary predictions
4. Evaluate Model
   python evaluate_model.py
   Generates comprehensive evaluation metrics and visualizations
5. 🌐 Web Interface (Recommended)
   streamlit run app.py
   Opens interactive web application at http://localhost:8501
   Features:
   • Modern gradient UI design
   • Interactive sliders and dropdowns
   • Real-time salary predictions
   • Percentile and comparison metrics
   • Interactive Plotly visualizations
   • Responsive layout

📈 Model Performance

• Linear Regression: R² = 0.30, MSE = 1923.22
• Polynomial Regression: R² = 0.33, MSE = 1851.86 ✅ (Best Model)

📁 Project Structure

archive (3)/
├── Salary.csv              # Dataset
├── eda.py                  # Exploratory data analysis
├── train_model.py          # Model training pipeline
├── train_model_advanced.py # Advanced training (7 models)
├── evaluate_model.py       # Model evaluation
├── inference.py            # Prediction script
├── app.py                  # 🌐 Web interface (Streamlit)
├── requirements.txt        # Dependencies
├── README.md              # Documentation
└── best_model.pkl         # Saved model

🎯 Features

• Comprehensive EDA with multiple visualizations
• Multiple regression algorithms comparison
• Automated best model selection
• Interactive prediction interface
• Model persistence with joblib
• Input validation and error handling

📝 Example Prediction

🔧 Technologies Used

• pandas: Data manipulation
• numpy: Numerical computations
• matplotlib & seaborn: Data visualization
• scikit-learn: Machine learning models
• joblib: Model serialization