Get / Delete Ground Truth Dataset

Retrieve a golden sample dataset with every expected value, or delete it permanently — deletion cascades to its validation runs. GET read, DELETE write scope.

GET /v1/validation/ground-truth/{id} retrieves the full detail of a golden sample (Ground Truth dataset), including every manually verified expected value entry; DELETE on the same path permanently removes it. These datasets are the reference side of Benchmarks: accuracy measurement of extraction output, distinct from the in-pipeline checks that gate results before delivery.

Call GET before starting a validation run to verify that expected values are correct and complete. The values array contains every document-field pair with its expected_value, document_id, and field_name, ordered oldest-first. Review these to ensure the benchmark data reflects your current extraction requirements — a stale expected value shows up later as a false mismatch in run results, and a systematically wrong one silently caps the best score any run can achieve.

The values array is filtered by document visibility: if source IAM rules hide a document from the user who minted your API key, that document's expected values are omitted from the response. user_schema_id confirms which schema the dataset is bound to (it is always set). entry_count on the detail response is null; values.length is the count after visibility filtering, and the list endpoint reports the unfiltered total.

Deleting a golden sample is destructive beyond the dataset itself: validation runs that reference the dataset are removed with it (the reference cascades), along with their per-field results. Export or record any scores you need before deleting. To update individual entries, recreate the dataset in the Talonic platform with corrected values and point new validation runs at it.

curl https://api.talonic.com/v1/validation/ground-truth/a1b2c3d4-e5f6-7890-abcd-ef1234567890 \
  -H "Authorization: Bearer tlnc_..."

Response

{
  "id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
  "name": "Invoice Validation Set",
  "user_schema_id": "b2c3d4e5-f6a7-8901-bcde-f12345678901",
  "entry_count": null,
  "created_at": "2024-08-01T00:00:00.000Z",
  "updated_at": "2024-08-01T00:00:00.000Z",
  "links": {
    "self": "/v1/validation/ground-truth/a1b2c3d4-e5f6-7890-abcd-ef1234567890"
  },
  "values": [
    {
      "id": "c3d4e5f6-a7b8-9012-cdef-123456789012",
      "document_id": "d4e5f6a7-b8c9-0123-defa-234567890123",
      "field_name": "invoice_number",
      "expected_value": "INV-2024-0042",
      "created_at": "2024-08-01T00:00:00.000Z"
    }
  ]
}

{
  "deleted": true
}

Errors