Normal score

This task applies a normal-score transformation to a continuous attribute. Use the method parameter to choose the direction:

• forward — transforms data from the source distribution to a standard normal (Gaussian) distribution. This is useful as a pre-processing step for geostatistical methods that require Gaussian input data.
• backward — transforms data from a standard normal distribution back to the source distribution. Use this to recover results in the original domain after Gaussian-based geostatistical computation.

Both directions preserve the rank order of the values.

Parameters

• method (string)
  • The direction of the transformation. Must be either "forward" or "backward".
• source (object)
  • object (geoscience object reference)
    • Reference to a geoscience object containing the values to transform. Must be a pointset, downhole-intervals, regular-3d-grid, regular-masked-3d-grid, or tensor-3d-grid.
  • attribute (geoscience object attribute reference)
    • Reference to a one-dimensional continuous attribute on the source object.

  • {
      "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-pointset.json",
      "attribute": "locations.attributes[?name=='my-data']"
    }

• distribution (geoscience object reference)
  • Reference to a non-parametric-continuous-cumulative-distribution object that defines the source distribution. Typically produced by the continuous-distribution task.
  • For a forward transformation this is the distribution of the source values.
  • For a backward transformation this is the distribution the result will follow.
• target (object)
  • object (geoscience object reference)
    • Reference to the geoscience object where the result will be stored. Must be a pointset, downhole-intervals, regular-3d-grid, regular-masked-3d-grid, or tensor-3d-grid.
  • attribute (geoscience object attribute target)
    • The attribute on the target object where the transformed values will be saved.

  • {
      "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-pointset.json",
      "attribute": {
        "operation": "create",
        "name": "my-normal-score-result"
      }
    }

Example

For more information, see the normal score API reference.

Forward transform request

requests.post(
    "https://{hub}.api.seequent.com/compute/orgs/{org_id}/geostatistics/normal-score",
    headers={"Authorization": "Bearer {token}"},
    json={
        "method": "forward",
        "source": {
            "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-pointset.json",
            "attribute": "locations.attributes[?name=='grade']",
        },
        "distribution": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-distribution.json",
        "target": {
            "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-pointset.json",
            "attribute": {
                "operation": "create",
                "name": "grade-gaussian",
            },
        },
    },
)

Backward transform request

requests.post(
    "https://{hub}.api.seequent.com/compute/orgs/{org_id}/geostatistics/normal-score",
    headers={"Authorization": "Bearer {token}"},
    json={
        "method": "backward",
        "source": {
            "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-kriged-grid.json",
            "attribute": "cell_attributes[?name=='grade-kriged-gaussian']",
        },
        "distribution": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-distribution.json",
        "target": {
            "object": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-kriged-grid.json",
            "attribute": {
                "operation": "create",
                "name": "grade-kriged",
            },
        },
    },
)

Result

{
    "message": "Normal score forward transformation completed.",
    "target": {
        "reference": "https://{hub}.api.seequent.com/geoscience-object/orgs/{org_id}/workspaces/{workspace_id}/objects/path/my-pointset.json",
        "name": "My Pointset",
        "description": null,
        "schema_id": "/objects/pointset/1.3.0",
        "attribute": {
            "reference": "locations.attributes[?key=='3cc3aa58-c928-4e79-b8ca-550174bff59e']",
            "name": "grade-gaussian"
        }
    }
}

Tips

• The source and target can be the same object, but they don't have to be.
• Pair the forward transform with the continuous-distribution task to build the reference distribution from your data before running the transformation.
• A common workflow is: forward transform → geostatistical estimation/simulation on Gaussian values → backward transform to recover results in the original domain.