Bagging Ensemble Studio 🚀

A robust, interactive Streamlit web application that allows users to build, configure, and evaluate Bagging Ensembles on their own datasets without writing any code.

🌟 Features

• Custom Data Upload: Upload any clean, preprocessed CSV dataset.
• Dynamic Configuration: Automatically detects columns to let you easily select your target and features.
• Task Versatility: Supports both Classification and Regression tasks.
• Base Estimator Selection: Choose and tune hyperparameters for various base models:
  • Decision Trees (Classifier/Regressor)
  • Linear/Logistic Regression
  • Ridge / Lasso
  • Support Vector Machines (SVC/SVR)
• Bagging Parameters: Fine-tune the ensemble by adjusting the number of estimators, max samples, max features, and bootstrapping.
• Validation Strategies: Evaluate your model using either a standard Train-Test Split or K-Fold Cross Validation.

🛠️ Installation

First, clone this repository to your local machine and navigate into the directory:

git clone https://github.com/yourusername/bagging-ensemble-studio.git
cd bagging-ensemble-studio

Choose one of the following methods to install the dependencies:

Option 1: Standard Installation using pip

It is recommended to set up a virtual environment first, then use the provided requirements.txt file:

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

# Install dependencies
pip install -r requirements.txt

Option 2: Ultra-fast Installation using uv

If you use uv (an extremely fast Python package installer and resolver), you can initialize a project environment and add the dependencies directly:

# Initialize a uv project
uv init

# Add the required packages
uv add streamlit pandas numpy scikit-learn

(Note: You can also simply run uv pip install -r requirements.txt to use the existing file with uv).

🚀 Usage

Run the Streamlit application from your terminal:

streamlit run app.py

This will open a new tab in your default web browser where you can interact with the Bagging Ensemble Studio.

📝 How to use the App

1. Upload Data: Use the file uploader to provide your clean .csv file.
2. Configure Data: Select if your task is Classification or Regression, then define your Target variable and Feature columns.
3. Set Base Algorithm: Choose your underlying model and tweak its specific hyperparameters (like Max Depth, C value, or Kernel).
4. Set Bagging Config: Adjust how the bagging meta-estimator behaves.
5. Evaluate: Choose your validation method and click "Train & Evaluate Ensemble" to see your metrics!