1
Write the evidence slice
Record baseline, candidate metric, sample context, one error slice, and one prohibited use.
Required evidence: A compact model-card table and decision paragraph.
0/80 minimum characters
Studio hour·45 minutes
Studio Hour: Model Card Slice
Turn one model comparison into a compact, portfolio-ready model-card section.
Scenario
A reviewer has only one page to decide whether your selected classifier is supported by the evidence and where it should not be used.
You will demonstrate
- Summarize comparative evidence.
- Name a concrete limitation.
- Write a bounded recommendation.
Project evidence
Show the work, not a checked box.
Each response is stored in this browser as you type. Include metrics, test output, or a decision rationale wherever the deliverable asks for it.
Runnable Python lab
Calculate one decision-aware slice
Compute precision for a fixed error slice before writing the model-card claim.
Project defense
What makes a model-card claim defensible?
Rubric self-review
Rate the evidence, not your effort: 0 missing, 1 weak, 2 adequate, 3 strong. All criteria must be reviewed, but a low honest score does not get hidden.
The recommendation cites a baseline, model evidence, a limitation, and a boundary.
Useful references
Project completion gate
Completion is controlled by stored evidence, deterministic tests, a decision defense, rubric review, and the artifact when required.
Evidence pendingCode pendingDefense pendingRubric pending
Optional cloud portfolio
Submit evidence across devices.
An account is required. Submit only when the local completion gate passes. AI review is advisory and separate from deterministic completion.