Sample-data demonstration
This shows our AI testing process on public sample data. You can't upload your own data here.
From YourCloudGroup Model Manager.

Question: How much demand you will have tomorrow
Sample data: Bike Sharing (UCI)
Attribution: Fanaee-T, H. (2013). Bike Sharing [Dataset]. UCI Machine Learning Repository. https://doi.org/10.24432/C5W894. CC BY 4.0. Modified: split into training and held-out rows, predictions added.
Predicts: cnt (rentals) (Regression)

Result
Its predictive value is MODERATE, measured on held-out rows that took no part in choosing it.
Certification: CERTIFIED — AUDITABLE & REPEATABLE PREDICTIVE MODEL
Value verified: yes. The winning model demonstrated useful predictive value over the baseline on holdout rows that took no part in choosing it, with limitations.
Winner: Basic NN, 10 epochs, best of 12 models built
Trained on: 629 rows

Certificate figures: the holdout half of the blind rows, 51 rows that took no part in choosing the winner
Error reduction: 22.8% (90% range 18.1-27.5%)
Mean absolute error: 1535.0 vs 1988.0 for the baseline
Baseline: always predict the training median (4,359)

Compare figures: all 102 blind rows, including the half the winner was chosen on
R2: 0.0542
MAE / RMSE: 1534.32 / 1928.99
Within 10%: 32 of 102

The predictions file
102 held-out rows, one per line: heldout_row, season, yr, mnth, holiday, weekday, workingday, actual_cnt, predicted_cnt.
heldout_row numbers the held-out rows 1 to N in the order the product tested them. It is not a row number from the original dataset: the split shuffled the rows.

Run: model 50e4d648-e8a8-4c41-a549-22e9f41a4fa2, tournament bike_day_calendar_20261028-213320, completed 2026-10-28T21:34:12Z
Dataset file sha256: 2c7649460e3cf1b6a96ee0b1e5864c83d9db746e3850b6071e0494275099aaf9