Datasets

Tip

Please consult our Documentation to create a dataset from the UI.

Please consult our full Swagger API documentation for more details on our APIs.

An organization can contain multiple Datasets associated to it. These are a good way to run bulk evaluations and collaborate with the Subject Matter Experts (SMEs) that typically annotate your data entries with high quality human labels and scores.

Create a Dataset

Creates a dataset within the defaut organization. Make sure to set your default organization right before this API request.

Example Request

curl -X 'POST' \
  'https://platform.lynxius.ai/api/datasets/' \
  -H "Authorization: Bearer $JWT_ACCESS_TOKEN" \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "name": "YOUR_DATASET_NAME"
}'

Returns

{
  "uuid": "f3fac660-cc27-42cd-bba0-89e9bdbaa7d4",
  "name": "YOUR_DATASET_NAME",
  "date_created": "2024-05-16T19:25:48.754Z",
  "organization_uuid": "f3fac660-cc27-42cd-bba0-89e9bdbaa7d4",
  "organization_name": "YOUR_ORGANIZATION_NAME"
}

Add a Dataset Entry

Adds an entry to the dataset specified by dataset_uuid.

Example Request

curl -X 'POST' \
  'https://platform.lynxius.ai/api/datasets/{dataset_uuid}/entries/' \
  -H "Authorization: Bearer $JWT_ACCESS_TOKEN" \
  -H 'accept: application/json' \
  -H 'Content-Type: application/json' \
  -d '{
  "query": "What is the main pizza ingredient in one word?",
  "output": "string",
  "reference": "Dough.",
  "score": "1",
  "comments": "Correct and answers in one word as requested."
}'

Returns

{
    "uuid": "cef9590c-a002-4cda-bb0a-f09b9b93deac"
}