Get Evaluation Stats | Humanloop Docs

from humanloop import Humanloop
client = Humanloop(api_key="YOUR_API_KEY", )
client.evaluations.get_stats(id='id', )

{
  "run_stats": [
    {
      "run_id": "run_id",
      "num_logs": 1,
      "evaluator_stats": [
        {
          "evaluator_version_id": "evaluator_version_id",
          "total_logs": 1,
          "num_judgments": 1,
          "num_nulls": 1,
          "num_errors": 1,
          "mean": 0,
          "sum": 0,
          "std": 1,
          "percentiles": {
            "0": -2.5,
            "25": -0.6745,
            "50": 0,
            "75": 0.6745,
            "90": 1.5,
            "95": 1.8,
            "99": 2,
            "100": 2.5,
            "99.9": 2.2
          }
        }
      ],
      "status": "pending",
      "version_id": "version_id",
      "batch_id": "batch_id"
    }
  ],
  "status": "pending",
  "progress": "progress",
  "report": "report"
}

Get Evaluation Stats.

Retrieve aggregate stats for the specified Evaluation. This includes the number of generated Logs for each Run and the corresponding Evaluator statistics (such as the mean and percentiles).

Authentication

X-API-KEYstring

API Key authentication via header

Path parameters

idstringRequired

Unique identifier for Evaluation.

Response

Successful Response

run_statslist of objects

Stats for each Run in the Evaluation.

statusenum

The current status of the Evaluation.

Allowed values:

progressstring

A summary string report of the Evaluation’s progress you can print to the command line;helpful when integrating Evaluations with CI/CD.

reportstring

A summary string report of the Evaluation you can print to command line;helpful when integrating Evaluations with CI/CD.

Errors

422

Get Stats Evaluations ID Stats Get Request Unprocessable Entity Error