Binary Classification Workflow

Introduction

This project focuses on applying a binary classification model to two different datasets: Banknote Authentication and Chronic Kidney Disease. It encompasses the entire machine learning workflow from data import and preprocessing to model training, validation, and analysis.

Datasets

1. Banknote Authentication Dataset: UCI Machine Learning Repository
2. Chronic Kidney Disease Dataset: Kaggle

Installation

To run this project, you need Python installed on your machine along with the following libraries:

• Pandas
• NumPy
• Scikit-learn
• Matplotlib
• Seaborn

You can install these packages using pip:

pip install pandas numpy scikit-learn matplotlib seaborn

File Description

• binary_classification_workflow.py: Contains all functions for data preprocessing, model training, validation, and result display.
• main.ipynb: Jupyter Notebook demonstrating the application of the workflow to the datasets.

Usage

1. Clone the repository from GitLab : https://gitlab.imt-atlantique.fr/m21aouad/mini-projet-intro-ml.git
2. Download the datasets and place them in the project directory.
3. Run the Jupyter Notebook to see the workflow in action.

Functions Overview

• Data Preprocessing: Functions for cleaning, filling missing values, scaling, and normalizing data.
• Model Training and Validation: Includes functions for splitting data, handling categorical features, feature selection, and model performance comparison.
• Utility Functions: For tasks like checking skewness, identifying outliers, and visualizing data.

Acknowledgements

This project is part of the course "Intro ML" at IMT Atlantique.