keneandita/irisintel repository overview

⁠Project Summary

This repository contains a containerized version of Iris Intel, an interactive Streamlit web application for predicting the species of an Iris flower based on its physical features. The application uses multiple pre-trained machine learning models, including Logistic Regression, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Decision Tree, to generate predictions and confidence scores.

The platform allows users to input flower measurements via sliders, provides real-time predictions, and displays confidence scores for each class. Dockerization ensures the application runs consistently without requiring local setup of Python, libraries, or model files.

⁠Architecture Overview

Iris Intel follows a modular architecture combining a Streamlit front-end interface and pre-trained machine learning models:

• Front-End Interface: Built with Streamlit, the UI provides interactive sliders for sepal and petal measurements and displays prediction results with confidence scores.
• Model Layer: Pre-trained models are stored as .joblib files in the Exported Models/ folder. Users can select which model to use for prediction.
• Prediction Pipeline: Inputs from the sliders are passed to the selected model, which returns a predicted class and confidence scores for each species.
• Container Environment: The Docker container packages Python, Streamlit, and required ML dependencies, along with model files, ensuring reproducible execution across systems.

⁠Key Features

Multiple ML Models

Supports Logistic Regression, KNN, SVM, and Decision Tree for prediction.

Interactive Streamlit Interface

Sliders allow users to input sepal and petal measurements, and predictions are updated in real time.

Confidence Scores

Displays prediction confidence percentages for all three Iris species: Setosa, Versicolor, and Virginica.

Pre-Trained Models

All models are pre-trained and included as .joblib files for immediate usage.

Clean and Responsive UI

Provides a user-friendly interface suitable for both desktop and mobile devices.

Containerized Deployment

The Docker image ensures consistent deployment and eliminates the need for local dependency installation.

⁠How to Run

First, pull the Docker image:

docker pull keneandita/irisintel

Run the container while exposing the Streamlit port:

docker run -p 8501:8501 keneandita/irisintel

Once the container starts, open the web app in a browser:

http://localhost:8501

Use the interactive sliders to input sepal and petal measurements. The app will display the predicted Iris species along with confidence scores for each class.