Get experiment by id | Opik Documentation

import requests
url = "http://localhost:5173/api/v1/private/experiments/id"
payload = {}
headers = {"Content-Type": "application/json"}
response = requests.get(url, json=payload, headers=headers)
print(response.json())

{
  "dataset_name": "Customer Churn Dataset",
  "id": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
  "dataset_id": "7c9e6679-7425-40de-944b-e07fc1f90ae7",
  "project_id": "1b4e28ba-2fa1-11d2-883f-0016d3cca427",
  "project_name": "Churn Prediction Project",
  "name": "Experiment 42 - Random Forest",
  "metadata": {
    "model_type": "random_forest",
    "hyperparameters": {
      "n_estimators": 100,
      "max_depth": 10
    }
  },
  "tags": [
    "random-forest",
    "churn",
    "2024"
  ],
  "type": "regular",
  "evaluation_method": "dataset",
  "optimization_id": "9f8c7d6e-1234-4b56-8a9b-0c1d2e3f4a5b",
  "feedback_scores": [
    {
      "name": "accuracy",
      "value": 0.87
    }
  ],
  "comments": [
    {
      "text": "Initial run with default parameters.",
      "id": "d290f1ee-6c54-4b01-90e6-d701748f0851",
      "created_at": "2024-01-15T09:30:00Z",
      "last_updated_at": "2024-01-15T09:30:00Z",
      "created_by": "alice@example.com",
      "last_updated_by": "alice@example.com"
    }
  ],
  "trace_count": 150,
  "dataset_item_count": 10000,
  "created_at": "2024-01-15T09:30:00Z",
  "duration": {
    "p50": 120.5,
    "p90": 180,
    "p99": 240
  },
  "total_estimated_cost": 35.75,
  "total_estimated_cost_avg": 0.24,
  "usage": {
    "cpu_hours": 12.5,
    "gpu_hours": 0,
    "memory_gb_hours": 50
  },
  "last_updated_at": "2024-02-01T12:00:00Z",
  "created_by": "alice@example.com",
  "last_updated_by": "bob@example.com",
  "status": "running",
  "experiment_scores": [
    {
      "name": "f1_score",
      "value": 0.82
    }
  ],
  "prompt_version": {
    "prompt_version_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "commit": "4f5e6d7c8b9a0e1f2d3c4b5a6e7f8d9c0b1a2c3d",
    "prompt_id": "123e4567-e89b-12d3-a456-426614174000",
    "prompt_name": "Initial Model Prompt"
  },
  "prompt_versions": [
    {
      "prompt_version_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
      "commit": "4f5e6d7c8b9a0e1f2d3c4b5a6e7f8d9c0b1a2c3d",
      "prompt_id": "123e4567-e89b-12d3-a456-426614174000",
      "prompt_name": "Initial Model Prompt"
    },
    {
      "prompt_version_id": "b2c3d4e5-f678-90ab-cdef-1234567890ab",
      "commit": "9a8b7c6d5e4f3a2b1c0d9e8f7a6b5c4d3e2f1a0b",
      "prompt_id": "223e4567-e89b-12d3-a456-426614174001",
      "prompt_name": "Tuned Model Prompt"
    }
  ],
  "dataset_version_id": "5f6e7d8c-9a0b-1c2d-3e4f-567890abcdef",
  "dataset_version_summary": {
    "id": "5f6e7d8c-9a0b-1c2d-3e4f-567890abcdef",
    "version_hash": "a1b2c3d4e5f67890abcdef1234567890",
    "version_name": "v3",
    "change_description": "Added new customer demographic features",
    "tags": [
      "demographics",
      "feature-update"
    ]
  },
  "pass_rate": 0.95,
  "passed_count": 950,
  "total_count": 1000
}

Get experiment by id

Path parameters

idstringRequiredformat: "uuid"

Response

Experiment resource

dataset_namestring or null>=1 character

idstring or nullformat: "uuid"

dataset_idstring or nullRead-onlyformat: "uuid"

project_idstring or nullformat: "uuid"

Project ID. Takes precedence over project_name when both are provided.

project_namestring or nullformat: "(?s)^\s*(\S.*\S|\S)\s*$"

Project name. Creates project if it doesn’t exist. Ignored when project_id is provided.

namestring or null

metadatamap from strings to any or list of objects or string or null

tagslist of strings or null

typeenum or null

Allowed values:

evaluation_methodenum or null

Allowed values:

optimization_idstring or nullformat: "uuid"

feedback_scoreslist of objects or nullRead-only

commentslist of objects or nullRead-only

trace_countlong or nullRead-only

dataset_item_countlong or nullRead-only

created_atdatetime or nullRead-only

durationobject or null

total_estimated_costdouble or nullRead-only

total_estimated_cost_avgdouble or nullRead-only

usagemap from strings to doubles or nullRead-only

last_updated_atdatetime or nullRead-only

created_bystring or nullRead-only

last_updated_bystring or nullRead-only

statusenum or null

Allowed values:

experiment_scoreslist of objects or null

prompt_versionobject or null

prompt_versionslist of objects or null

dataset_version_idstring or nullformat: "uuid"

ID of the dataset version this experiment is linked to. If not provided at creation, experiment will be automatically linked to the latest version.

dataset_version_summaryobject or nullRead-only

Summary of the latest dataset version

pass_ratedouble or nullRead-only

Pass rate for evaluation suite experiments (0.0-1.0). Null for regular experiments.

passed_countlong or nullRead-only

Number of items that passed for evaluation suite experiments. Null for regular experiments.

total_countlong or nullRead-only

Total number of items for evaluation suite experiments. Null for regular experiments.

Path parameters

Response

Errors