Palmer Penguin Classification
A Streamlit machine learning app that predicts Palmer penguin species from user-provided physical measurements.
Project Overview
An interactive machine learning web app built with Python and Streamlit to classify Palmer penguins as Adelie, Chinstrap, or Gentoo. The app allows users to either upload a CSV file or manually enter penguin features such as island, sex, bill length, bill depth, flipper length, and body mass. It preprocesses the input using one-hot encoding, loads a saved classification model, and displays both the predicted species and prediction probabilities.
The Problem
Machine learning models are often difficult for non-technical users to interact with because they usually require code, notebooks, or command-line execution. For a classification task like Palmer penguin species prediction, users need a simple interface where they can enter measurements and instantly see predictions.
The Solution
Built a Streamlit application that collects penguin measurements through sidebar controls or CSV upload, preprocesses categorical variables using one-hot encoding, loads a saved scikit-learn classification model, and returns the predicted penguin species with class probability scores.
The Impact
Created a simple, accessible machine learning demo that turns a trained classification model into an interactive web app, making it easier to showcase model deployment, user input handling, preprocessing, and prediction workflows.
Date / Year
2022