Get Dataset#

Prerequisites#

Before you can get an existing dataset, make sure that you have:

  • The namespace and dataset name for the dataset you want to retrieve.


How to Get a Dataset#

API#

You can send requests to the NeMo Entity Store Microservice to get a dataset object.

  1. Make a GET request to the /v1/datasets/{namespace}/{dataset_name} endpoint.

    export ENTITY_STORE_BASE_URL=<URL for NeMo Entity Store>
    
    curl -X GET "${ENTITY_STORE_BASE_URL}/v1/datasets/${NAMESPACE}/${DATASET_NAME}" \
        -H 'Accept: application/json' \
        -H 'Content-Type: application/json' | jq
    
  2. Check the dataset’s details by reviewing the response.

    Example Response
    {
      "schema_version": "1.0",
      "id": "dataset-81RSQp7FKX3rdBtKvF9Skn",
      "description": "A dataset for documentation testing",
      "type_prefix": null,
      "namespace": "team-docs",
      "project": "string",
      "created_at": "2025-02-14T20:47:20.798490",
      "updated_at": "2025-02-14T20:47:20.798492",
      "custom_fields": {},
      "ownership": {
        "created_by": "user@nvidia.com",
        "access_policies": {}
      },
      "name": "documentation-test-dataset",
      "version_id": "main",
      "version_tags": [],
      "format": "json",
      "files_url": "file://documentation-test-dataset.json/"
    }