Iris Species Classification and Visualization Web App

Overview

This project is an interactive web application built using Streamlit for Iris species classification and data visualization. It demonstrates the end-to-end workflow of a machine learning project, including data preprocessing, model training, and deployment. Additionally, it collects user input and stores it in a database, showcasing backend integration.

---

Features

• User Interaction:

  • Collects user details (name, email, age).
  • Allows users to input Iris flower measurements for predictions.

• Iris Species Prediction:

  • Utilizes a Random Forest Classifier to predict Iris species (Setosa, Versicolor, Virginica).

• Data Visualization:

  • Feature correlations using a heatmap.
  • Principal Component Analysis (PCA) for dimensionality reduction and visualization.

• Model Performance:

  • Provides an option to train the model and evaluate its performance.
  • Displays the classification report and confusion matrix.

• Database Integration:

  • Stores user information (name, email, age, and prediction data) in a SQLite database.

---

Technologies Used

• Python Libraries:
  • Streamlit, Pandas, Scikit-learn, Matplotlib, Seaborn, Shap, SQLite3
• Machine Learning:
  • Random Forest Classifier
• Visualization:
  • Heatmaps, PCA Scatter Plots
• Backend:
  • SQLite database for storing user data

---

Setup Instructions

1. Clone the repository:

   git clone https://github.com/vishwaspw/streamlit-iris-classifier.git
   cd streamlit-iris-classifier