Model Manager · how AI testing works

Sample-data demonstration

This shows our AI testing process on public sample data. You can't upload your own data here.

  1. 1

    Pick a sample dataset

    Public datasets, each a real business question.

  2. 2

    See how AI tested it

    The same process, already run on that data.

  3. 3

    Download predictions and results

    Every held-back row with its actual outcome and the prediction.

Step 1

Pick a sample dataset

Which customers are about to leave

Iranian Churn (UCI) · Classification · 272 held-out rows

Who will say yes to an offer

Bank Marketing (UCI) · Classification · 4,518 held-out rows

What a property will sell for

Real Estate Valuation (UCI) · Regression · 103 held-out rows

Which applicants are a higher credit risk

Statlog German Credit (UCI) · Classification · 100 held-out rows

Which website visitors will buy

Online Shoppers Purchasing Intention (UCI) · Classification · 1,215 held-out rows

How much demand you will have tomorrow

Bike Sharing (UCI) · Regression · 102 held-out rows

Which products will be rated highly

Wine Quality (UCI) · Classification · 322 held-out rows

Each sample is a public dataset from the UCI Machine Learning Repository under CC BY 4.0; the attribution is on its page.