# Hybrid Support Vector Regression Model with Optimization Algorithms

This repository contains Python code for optimizing Support Vector Regression (SVR) model hyperparameters using various optimization algorithms. The purpose is to enhance the prediction accuracy of sIL-2R levels in patients with sarcoidosis based on their serum ACE levels. The hybrid optimization approach combines Bald Eagle Search (BES) and Chimp Optimizer (CO), while comparative models use Firefly Algorithm (FFA), Grey Wolf Optimization (GWO), and BES alone.

## Repository Structure

- `hybrid_optimization.py`: Contains the hybrid optimization methodology using BES and CO algorithms for SVR hyperparameter tuning.
- `comparative_models.py`: Contains the comparative models using FFA, GWO, and BES algorithms for SVR hyperparameter tuning.
- `README.md`: This file providing an overview and instructions for the repository.

## Installation

1. Clone the repository:
    ```sh
    git clone https://github.com/yourusername/svr-optimization.git
    cd svr-optimization
    ```

2. Create a virtual environment and activate it (optional but recommended):
    ```sh
    python -m venv env
    source env/bin/activate  # On Windows use `env\Scripts\activate`
    ```

3. Install the required packages:
    ```sh
    pip install numpy scikit-learn
    ```

4. Install optimization algorithm packages:
    ```sh
    pip install firefly-algorithm grey-wolf-optimizer bald-eagle-search
    ```
    Note: Replace with actual package names or custom implementations if they are not available.

## Usage

### Hybrid Optimization

The `hybrid_optimization.py` script implements a hybrid BES-CO optimization process for tuning SVR hyperparameters. To run the script:

1. Ensure you have the necessary data and packages installed.
2. Execute the script:
    ```sh
    python hybrid_optimization.py
    ```

This script will:
- Generate synthetic data.
- Initialize the population and perform BES and CO optimization phases.
- Train the SVR model using optimized hyperparameters.
- Evaluate and print the model performance.

### Comparative Models

The `comparative_models.py` script implements SVR hyperparameter tuning using FFA, GWO, and BES algorithms. To run the script:

1. Ensure you have the necessary data and packages installed.
2. Execute the script:
    ```sh
    python comparative_models.py
    ```

This script will:
- Generate synthetic data.
- Optimize SVR hyperparameters using FFA, GWO, and BES algorithms.
- Train the SVR models using optimized hyperparameters.
- Evaluate and print the performance of each model.