Create Evaluation

• Endpoint
• Request Body
• Request Example
• Response Example

Endpoint

POST /api/v1/evaluations

Create a new evaluation for the authenticated team.

Request Body

The request body must be a JSON object with the following properties:

• name (string) required: The name of the evaluation.
• description (string) optional: A description of the evaluation.
• model_id (integer) required: The ID of the search model to be evaluated.
• scale_type (string) required: The scale type of the evaluation. Possible values are binary, graded, detail.
• metrics (array of objects) required: A list of metrics in the format {"scorer_type": "ap", "num_results": 10}.
  • scorer_type (string) required: The type of scorer. Possible values are precision, ap, rr, cg, dcg, ndcg, cg_d, dcg_d, ndcg_d.
  • num_results (integer) required: The number of results to be evaluated.
• keywords (array of strings) required: A list of keywords for the evaluation.
• tags (array of objects) optional: A list of tags for the evaluation in the format {"id": 1}.
  • id (integer) required: The ID of the tag.
• setting_feedback_strategy (integer) required: The feedback strategy. Possible values are 1 (Single), 3 (Multiple).
• setting_show_position (boolean) required: Whether to show the position of the results.
• setting_reuse_strategy (integer) required: The reuse strategy for grades. Possible values are 0 (No), 1 (Query/Doc), 2 (Query/Doc/Position).
• setting_auto_restart (boolean) required: Whether to auto-restart the evaluation.
• setting_scoring_guidelines (string) optional: The scoring guidelines for the evaluation. You can use Markdown syntax in this field. For more information, see the Markdown Guide
• transformers (object) required: The transformers for the evaluation.
  • scale_type (string) required: The source scale type of the transformers. Possible values are binary, graded, detail. Must match the scale_type of the evaluation.
  • rules (object) required: The rules for the transformers. Must contain rules for each required scale type. If transformers are not required, set to an empty object.

Request Example

curl -X POST "https://searchtweak.com/api/v1/evaluations" \ 
    -H "Authorization: Bearer YOUR_API_TOKEN" \
    -H "Accept: application/json" \
    -H "Content-Type: application/json" \
    -d '{
          "model_id": 1,
          "name": "Sample Evaluation",
          "description": "This is a sample evaluation.",
          "keywords": [
            "apple",
            "dyson"
          ],
          "scale_type": "detail",
          "metrics": [
            { "scorer_type": "precision", "num_results": 10 },
            { "scorer_type": "ndcg", "num_results": 10 },
            { "scorer_type": "ndcg_d", "num_results": 5 }
          ],
          "transformers": {
            "scale_type": "detail",
            "rules": {
              "binary": {
                "1": 0,
                "2": 0,
                "3": 0,
                "4": 0,
                "5": 0,
                "6": 1,
                "7": 1,
                "8": 1,
                "9": 1,
                "10": 1
              },
              "graded": {
                "1": 0,
                "2": 1,
                "3": 1,
                "4": 1,
                "5": 2,
                "6": 2,
                "7": 2,
                "8": 3,
                "9": 3,
                "10": 3
              }
            }
          },
          "tags": [
            { "id": 9 }
          ],
          "setting_feedback_strategy": 1,
          "setting_reuse_strategy": 0,
          "setting_show_position": true,
          "setting_auto_restart": false,
          "setting_scoring_guidelines": "Scoring guidelines for the evaluation. You can use **Markdown** syntax in this field."
        }'

Response Example

{
  "id": 307,
  "model_id": 1,
  "scale_type": "detail",
  "status": "pending",
  "progress": 0,
  "name": "Sample Evaluation",
  "description": "This is a sample evaluation.",
  "settings": {
    "reuse": 0,
    "position": true,
    "strategy": 1,
    "auto_restart": false,
    "transformers": {
      "rules": {
        "binary": {
          "1": 0,
          "2": 0,
          "3": 0,
          "4": 0,
          "5": 0,
          "6": 1,
          "7": 1,
          "8": 1,
          "9": 1,
          "10": 1
        },
        "graded": {
          "1": 0,
          "2": 1,
          "3": 1,
          "4": 1,
          "5": 2,
          "6": 2,
          "7": 2,
          "8": 3,
          "9": 3,
          "10": 3
        }
      },
      "scale_type": "detail"
    },
    "setting_scoring_guidelines": "Scoring guidelines for the evaluation. You can use **Markdown** syntax in this field."
  },
  "metrics": [
    {
      "scorer_type": "precision",
      "num_results": 10,
      "value": null
    },
    {
      "scorer_type": "ndcg",
      "num_results": 10,
      "value": null
    },
    {
      "scorer_type": "ndcg_d",
      "num_results": 5,
      "value": null
    }
  ],
  "tags": [
    {
      "id": 9,
      "name": "All"
    }
  ],
  "keywords": [
    "apple",
    "dyson"
  ],
  "created_at": "2024-11-03T08:05:39+00:00",
  "finished_at": null
}