Evaluations
Choisissez votre langage :
Evaluations run Mankinds tests against a system.
client.evaluations.run(params)
Start an evaluation run.
| SDK | client.evaluations.run(params, idempotency_key=None) |
|---|---|
| Retourne | EvaluationRun |
Parameters:
| Emplacement | Paramètre | Type | Requis | Défaut | Description |
|---|---|---|---|---|---|
| Body | system_id | string | Requis | — | System UUID. |
| Body | profile | string | Optionnel | — | Evaluation profile. |
| Body | connection_id | string | Optionnel | — | Connection UUID. Uses the system default when omitted. |
| Body | mode | string | Optionnel | offline | Evaluation mode. |
| Body | thematics_config | dict | Optionnel | — | Offline criteria/thematics configuration. |
| Body | online_thematics_config | dict | Optionnel | — | Online criteria/thematics configuration. |
| Body | sampling | dict | Optionnel | — | Sampling configuration. |
| Body | time_range | dict | Optionnel | — | Time range for online/mixed evaluations. |
| Body | connector_refs | dict | Optionnel | — | External connector references. Defaults to an empty object. |
| SDK option | idempotency_key | string | Optionnel | — | Sent as the Idempotency-Key header. |
The response includes evaluation_id, status, task_id, job and message. Use evaluation_id with client.evaluations.get(...) or client.evaluations.wait(...).
client.evaluations.list(filters)
List evaluations for the organization or for one system.
| SDK | client.evaluations.list(filters=None) |
|---|---|
| Retourne | Evaluation[] |
Parameters:
| Emplacement | Paramètre | Type | Requis | Défaut | Description |
|---|---|---|---|---|---|
| Query | system_id | string | Optionnel | — | Restrict results to one system. |
| Query | limit | integer | Optionnel | — | Maximum number of evaluations. Minimum 1, maximum 200. |
| Query | offset | integer | Optionnel | 0 | Pagination offset. |
client.evaluations.get(evaluationId)
Read one evaluation.
| SDK | client.evaluations.get(evaluation_id) |
|---|---|
| Retourne | Evaluation |
Parameters:
| Emplacement | Paramètre | Type | Requis | Description |
|---|---|---|---|---|
| Path | evaluation_id | string | Requis | Evaluation UUID. |
client.evaluations.cancel(evaluationId)
Cancel a running evaluation.
| SDK | client.evaluations.cancel(evaluation_id) |
|---|---|
| Retourne | StatusResponse |
Parameters:
| Emplacement | Paramètre | Type | Requis | Description |
|---|---|---|---|---|
| Path | evaluation_id | string | Requis | Evaluation UUID. |
client.evaluations.wait(evaluationId, options)
SDK helper that polls client.evaluations.get(...) until the evaluation reaches a terminal state.
| SDK | client.evaluations.wait(evaluation_id, timeout_ms=..., interval_ms=..., on_poll=None) |
|---|---|
| Retourne | Evaluation |
Parameters:
| Emplacement | Paramètre | Type | Requis | Défaut | Description |
|---|---|---|---|---|---|
| SDK | evaluation_id | string | Requis | — | Evaluation UUID. |
| SDK | timeout_ms | integer | Optionnel | 1800000 | Maximum wait time in milliseconds. |
| SDK | interval_ms | integer | Optionnel | 5000 | Poll interval in milliseconds. |
| SDK | on_poll | callable | Optionnel | — | Callback called with each evaluation payload. |
Raises PollTimeoutError on timeout and JobFailedError if the terminal status is failed, canceled or timed out.
Cette page vous a été utile ?