Read-only demo No credits charged No private job access
Longitudinal Classification

Repeated-Visit Risk Stratification

How does repeated subject history change the risk category over time?

A longitudinal classification walkthrough showing subject ID, time variable, readiness review, mixed-effects model comparison, and model review notes.

Demo outcome
GPBoost Classifier
Best model
Selected from replayed saved metrics.
0.460
F1 Score
Primary metric for this workflow.
8 minutes
Observed run time
Demo replay based on saved job behavior.
This workspace is a guided public replay.
Training, SHAP, report generation, prediction, job deletion, and billing actions are unavailable in demo mode.
Completed Demo ID: demo-longitudinal-classification
Saved-output workflow

The demo, handbook, and job page use the same saved outputs.

Each item below comes from the completed source job, so the public replay matches what users review after a real run.

Open AI Report tab
Data setup
Summary Only
The example records the data roles and checks shown in the replay.
None
Analysis checks
Summary Only
The replay explains the checks expected for this type of analysis.
None
Model explanation
Summary Only
The replay describes how model findings should be interpreted.
None
Reproducibility boundary
Available
The demo is read-only and does not run training, prediction, or billing.
None
Methods appendix
Available
A concise methods note is bundled with this public example.
methods_appendix.md
Dataset and roles

Readiness preview input

Rows
2391
Features
8
Target
risk_group
Format
Parquet
Status
Pass with notes
subject id time target features
Workflow steps

How the job moved

Map repeated-measure roles
Subject and time columns were confirmed before readiness preview.
Apply runtime guardrails
Heavy mixed-effects models were checked against the interactive runtime budget.
Review repeated-measure outputs
Model metrics, confidence intervals, and guardrail notes are available.
Saved outputs

What users review

Role mapping
readiness
Subject ID, time, target, and feature columns.
Guardrail summary
runtime
Runtime-budget decisions for heavy longitudinal models.
Model review summary
review
Review status, reviewer notes, and follow-up decision.
Model comparison

Replayed performance table

F1 Score primary
Model Status Metric Value Training Time
GPBoost Classifier Best F1 Score 0.460 0.5s
Fixed Effects Logistic Trained F1 Score 0.422 0.5s
HCB Trained F1 Score 0.238 221s
REEM CatBoost Skipped by limit Runtime guardrail Skipped 0s
Report summary

Decision-ready notes

  • The demo shows how longitudinal roles are mapped before training.
  • The best model was selected from model performance artifacts.
  • Heavy-model guardrails should be interpreted as controlled skips, not failures.
Subject-aware split
51 test subjects

Evaluation preserves subject-level grouping where possible.

Model comparison
GPBoost selected

The best model is selected from saved metrics, not from SHAP output.

Guardrail note
1 model skipped

A runtime skip is not a training crash; it protects interactive execution.

Live log replay

The event stream users see during a real run

Press Replay logs to stream the curated demo events.
Ready for your own data?

Use the real workflow with authentication and billing safeguards.

A real run requires sign-in, readiness preview, visible cost estimate, and confirmation before training starts.