Your learning data is designed to stay local.

Last updated: 2026-08-31. This page describes the behavior of the application contained in this repository. Hosting providers can still receive ordinary web-request metadata when they deliver the static site.

Local datasets and model training

CSV files selected in the application are parsed in the browser. The application does not require a server-side ML API to train the included models, and the production host is designed to serve static HTML, CSS, JavaScript, worker files, and assets.

Browser storage

Learning progress, experiments, workspace notes, and related state may be stored in browser storage on your device so the application can restore your work. You can remove this data using the application's reset controls where available or by clearing site data in your browser.

Hosting logs

Like most websites, the static hosting provider or reverse proxy may process connection information such as IP address, request path, timestamp, user agent, and security/error logs. That infrastructure data is separate from the contents of local CSV files and model calculations.

Analytics, advertising, and accounts

The production build in this package does not include an advertising SDK, cross-site tracking script, cloud account system, or analytics service. If those are added later, this notice should be updated before deployment.

Permissions

The supplied production security policy disables camera, microphone, geolocation, payment, and USB permissions because the current application does not need them.

Questions and changes

This repository does not embed a public support email address. The site owner should add a verified contact method before collecting user enquiries or other personal information.