Skip to main content
Inspect value learner state
Inspect Adapt-1 value state with session_id. Recorded responses expose configuration, sample and dataset identities, installed model, candidate, training state, and the frozen flag. Use training.status to monitor a fit; top-level status can remain configured while training is running. The training guide explains report fields. The request tables below describe schema version 5. A default marked “Not declared” is absent from that schema; it does not mean null. Check /version at the API host root and confirm compatibility with your deployment.

Path, headers, and query

Send the hosted API key as Authorization: Bearer <ADAPT1_API_KEY>. The authenticated identity determines ownership; follow HTTP and contracts for the session placeholder. X-Neuroadapt-Session-Id is a compatibility header in schema version 5. It does not replace the hosted bearer identity.

JSON body contract

Content type: application/json. The HTTP request body is required. The following table describes its fields; linked schema names lead to the complete nested definitions on this page. When an array type also permits null, that alternative applies to the entire field, not to individual array items.

Request body

Schema: ValueRequest. Fields outside this object’s declared properties are rejected by the schema.

Responses

The validation error shape has an optional detail array. Each validation entry requires loc (an array of strings or integers), msg (string), and type (string); it can also include ctx (object) and input (any JSON value). Operational refusals can have their own error details. Recorded state reports include sample_count, dataset identities, configuration, model and candidate summaries, training, and frozen. Preserve the observation rows in your application record alongside these counts and dataset identities.

Monitor a training request

Keep the state returned before submission with your training request journal, then follow the polling procedure for the intended fit. Handle the following training statuses: This table covers reported training statuses. Resolve any additional status against your deployment’s contract before recording completion. Match training.request_id to the submitted request, then use that fit’s status, installation flag, and installed model identity to confirm its result. Bind prediction to the model you intend to evaluate with expected_model_sha256. Use the training-state guide for field interpretation and refusal diagnostics. A client deadline ends the client’s wait; reconcile state before another training write.

Match a result to your request and model

These nested fields describe observed response shapes. Keep the complete state response when a deployment returns a different shape. Complete a wait from the matching job’s status and use the candidate’s own hash for candidate evaluation. See the complete workflow client and prediction response checks.