Sensor measurement classes: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Use one sensor's description to distinguish supply consumption, stored contents and another volume measurement. Choose among stipulated unit-compatible classes without converting units or reading a live sensor.
Units can narrow the set of allowed classes without deciding among them. Water consumed from a pipe and water remaining in a tank can share a unit while representing different quantities.
This page deliberately uses a small stated choice set. A production integration should obtain allowed units from its own version of Home Assistant; this demo does not validate the entire registry.
This original text-metadata exercise adapts a question in ha-gutcheck, linked by the source post. The inputs and conservative review outcomes are authored here. It does not connect to Home Assistant, change a device registry or operate equipment.

Use the sample to understand the rule, then test your own input for sensor measurement classes.

Use one sensor's description to distinguish supply consumption, stored contents and another volume measurement.

Open the decision rules and keep the intended distinction between Water, Gas, Stored, Volume, Unclear. Edit the choices only when testing your own rule.

Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

Review Unclear before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Supply water | Water | 100% |
| Supply gas | Gas | 100% |
| Tank contents | Stored | 99.0% |
| Other volume | Volume | 97.0% |
| Generic name | Volume | 82.0% |
| Instruction in a name | Water | 100% |
These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 5 of 6 results matched the authored fixture expectations. For Generic name, the expectation was Unclear; the observed choice was Volume.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
In this contract Water means consumption from a supply. Stored means current container contents, regardless of whether the substance is water.
Not in these examples. Every offered class accepts the stipulated unit, so the description must establish what is measured.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.