Skip to content

Application to predict whether a squirrel in NYC's Central Park will approach you based on a random forest model

License

Notifications You must be signed in to change notification settings

victormurcia/SquirrelML

Repository files navigation

Predicting Squirrel Approach: Streamlit App

This Streamlit app is designed to predict the likelihood of a squirrel approaching based on various input features. Utilizing a machine learning model trained on comprehensive squirrel behavior data, the app provides an interactive interface for users to input data and receive predictions.

The app is here: https://squirrelml.streamlit.app/

Features

  • Interactive Map: Select geographical coordinates to determine the location of your squirrel observation.
  • Customizable Inputs: Adjust various parameters like time of day, squirrel age, fur color, and observed behaviors to refine your prediction.
  • Real-Time Predictions: Instantly receive a probability score indicating the likelihood of a squirrel's approach.
  • Model Insights: Understand the key factors influencing the prediction with an integrated machine learning explanation tool.

Using the App

  • Set Location: Click on the map or manually enter the latitude and longitude of your squirrel observation.
  • Input Parameters: Use the dropdowns and checkboxes to input details like the time of day, squirrel age, primary and highlight fur color, and various behaviors observed.
  • Get Prediction: Click on the 'Predict' button to view the likelihood of a squirrel approaching, presented as a percentage.

About

Application to predict whether a squirrel in NYC's Central Park will approach you based on a random forest model

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published