Retriever Evaluations

Get evaluation results

Changed on

Retrieve evaluation results with all calculated metrics

get/v1/retrievers/{retriever_id}/evaluations/{evaluation_id}

Request

  • Base URL: https://api.mixpeek.com
  • URL: https://api.mixpeek.com/v1/retrievers/{retriever_id}/evaluations/{evaluation_id}
  • Auth: HTTP bearer and API key in header X-Namespace

Path parameters

retriever_idstring required
evaluation_idstring required

Response

Successful Response

evaluation_idstring required

Unique evaluation identifier

retriever_idstring required

ID of retriever being evaluated

dataset_idstring required

ID of dataset used for evaluation

dataset_namestring required

Name of dataset

status'pending' | 'in_progress' | 'completed' | 'failed' required

Status of an evaluation run.

created_atstring date-time required

When evaluation was created

updated_atstring date-time required

Last update timestamp

completed_atstring date-time nullable

When evaluation completed

namespace_idstring required

Namespace ID

internal_idstring required

Internal organization ID

query_countinteger required

Number of queries evaluated

overall_metricsobject nullable

Aggregated metrics across all queries

metrics_by_kobject nullable

Metrics broken down by K value (keys are string K values like '5', '10', '20')

total_queriesinteger nullable

Total queries in the dataset for this run (= evaluated_queries + skipped_queries).

evaluated_queriesinteger nullable

Number of queries that produced metrics. May be < total_queries when some queries were skipped (skip-and-continue on empty/failing input).

skipped_queriesinteger nullable

Number of queries skipped during evaluation (empty query_input or a per-query execution failure) — these did not fail the whole eval.

error_messagestring nullable

Error message if failed

baseline_evaluation_idstring nullable

For a trained weight set's evaluation: the evaluation of the live configuration on the same dataset that this run was gated against.

weight_set_idstring nullable

For a trained weight set's evaluation: the weight set this run scored. When the verdict passed, the stage's learning_config.shipped_weight_set carries the same id.

Changes

  • 61a175e8eb9aRevision changes
    • warning

      removed the optional property from the response with the status

    • info

      added the non-success response with the status

    • info

      added the non-success response with the status

    • info

      added the optional property to the response with the status

    • info

      added the required property to the response with the status

    • info

      added the required property to the response with the status

  • b75474e978ffRevision changes
    • info

      added the optional property to the response with the status

    • info

      added the optional property to the response with the status

    • info

      added the optional property to the response with the status