/E2E-Customer-Segmentation-PyCaret-Streamlit

We build end-to-end unsupervised solution for customer segmentation using PyCaret deploy the model using Streamlit.

Primary LanguageJupyter NotebookMIT LicenseMIT

End to End Customer Segmentation using PyCaret and Streamlit

Business Objective

Customers are crucial for any organization, and customer retention is key for long-term success. This project focuses on customer segmentation leveraging predictive modeling, data visualization, and segmentation to optimize marketing efforts.

Data Description

The dataset, named 'jewellery,' includes customer profiles with attributes like age, income, spending score, and savings. We aim to classify customers into segments for targeted marketing.

Aim

Build an end-to-end unsupervised solution for customer segmentation using PyCaret to categorize customers into segments and deploy the model using Streamlit.

Tech Stack

  • Language: Python
  • Libraries: PyCaret, Pandas, Streamlit

Approach

  1. Import required libraries and packages
  2. Open the configuration file
  3. Get the dataset
  4. Setup PyCaret environment
  5. Model Creation
  6. Model Assigning
  7. Plotting model
  8. Making predictions
  9. Saving Model
  10. Creating Streamlit application
  11. Creating a GitHub repository for the project
  12. Connecting Streamlit Cloud to GitHub
  13. Deploying the project

Code Overview

  1. input

    • Config file
    • jewel_data.csv with customer data

  2. src

    • engine.py
    • ml_pipeline
      • Folder containing modularized code for various steps
    • streamlit_app
      • Folder with the Streamlit application file
    • requirements.txt for package installation

  3. output

    • Model trained on the data for future use

  4. lib

    • Reference folder containing the original ipython notebook