Evaluate a trained model

POST /_ml/trained_models/{model_id}/deployment/_infer
POST /_ml/trained_models/{model_id}/_infer

Auth

Path parameters

model_id string required: The unique identifier of the trained model.

Query parameters

timeout union oneOf · default: 10s: Controls the amount of time to wait for inference results.; Values: -1 0

Request body

docs string to object[] required: An array of objects to pass to the model for inference. The objects should contain a fields matching your configured trained model input. Typically, for NLP models, the field name is text_field. Currently, for NLP models, only a single value is allowed.
inference_config object: The inference configuration updates to apply on the API call

Response

application/json

inference_results object[] required

View as Markdown

Examples

Request Console 1 / 8 languages

An example body for a POST _ml/trained_models/lang_ident_model_1/_infer request.

Console Kibana Dev Tools

POST _ml/trained_models/lang_ident_model_1/_infer
{
  "docs":[{"text": "The fool doth think he is wise, but the wise man knows himself to be a fool."}]
}

JavaScript Elasticsearch JavaScript Client

const response = await client.ml.inferTrainedModel({
  model_id: "lang_ident_model_1",
  docs: [
    {
      text: "The fool doth think he is wise, but the wise man knows himself to be a fool.",
    },
  ],
});

Python Elasticsearch Python Client

resp = client.ml.infer_trained_model(
    model_id="lang_ident_model_1",
    docs=[
        {
            "text": "The fool doth think he is wise, but the wise man knows himself to be a fool."
        }
    ],
)

curl Shell

curl -X POST "$ELASTICSEARCH_URL/_ml/trained_models/lang_ident_model_1/_infer" \
  -H "Authorization: ApiKey $ELASTIC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "docs": [
      {
        "text": "The fool doth think he is wise, but the wise man knows himself to be a fool."
      }
    ]
  }'

Java Elasticsearch Java Client

client.ml().inferTrainedModel(i -> i
    .docs(Map.of("text", JsonData.fromJson("\"The fool doth think he is wise, but the wise man knows himself to be a fool.\"")))
    .modelId("lang_ident_model_1")
);

C# Elasticsearch .NET Client

var response = await client.MachineLearning
    .InferTrainedModelAsync(modelId: "lang_ident_model_1", d1 => d1
        .Docs([new Dictionary<string, object>()
        {
            { "text", "The fool doth think he is wise, but the wise man knows himself to be a fool." }
        }])
    );

PHP Elasticsearch PHP Client

$resp = $client->ml()->inferTrainedModel([
    "model_id" => "lang_ident_model_1",
    "body" => [
        "docs" => array(
            [
                "text" => "The fool doth think he is wise, but the wise man knows himself to be a fool.",
            ],
        ),
    ],
]);

Ruby Elasticsearch Ruby Client

response = client.ml.infer_trained_model(
  model_id: "lang_ident_model_1",
  body: {
    "docs": [
      {
        "text": "The fool doth think he is wise, but the wise man knows himself to be a fool."
      }
    ]
  }
)

Response

{
  "inference_results": [
    {
      "entities": [
        {
          "class_name": "string",
          "class_probability": 0,
          "entity": "string",
          "start_pos": 0,
          "end_pos": 0
        }
      ],
      "is_truncated": false,
      "predicted_value": 0,
      "predicted_value_sequence": "string",
      "prediction_probability": 0,
      "prediction_score": 0,
      "top_classes": [
        {
          "class_name": "string",
          "class_probability": 0,
          "class_score": 0
        }
      ],
      "warning": "string",
      "feature_importance": [
        {
          "feature_name": "string",
          "importance": 0,
          "classes": []
        }
      ]
    }
  ]
}

Generated from the API spec