Create and manage prediction results

This page shows you how to create and manage AML AI prediction results. Prediction results are saved to BigQuery tables.

You only need to create the prediction results and export the associated metadata at this point. The other prediction results methods are provided as a convenience.

Before you begin

To get the permissions that you need to create and manage prediction results, ask your administrator to grant you the Financial Services Admin (financialservices.admin) IAM role on your project. For more information about granting roles, see Manage access.

You might also be able to get the required permissions through custom roles or other predefined roles.
Create an instance
Create a model
Create a dataset

Create prediction results

Some API methods return a long-running operation (LRO). These methods are asynchronous. The operation might not be completed when the method returns a response. For these methods, send the request and then check for the result.

Send the request

To create prediction results, use the projects.locations.instances.predictionResults.create method.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: a user-defined identifier for the instance
PREDICTION_RESULTS_ID: a user-defined identifier for the prediction results
MODEL_ID: a user-defined identifier for the model
DATASET_ID: the user-defined identifier for the dataset used for predictions; tables should not have the training label columns
PREDICTION_END_DATE: The latest time from which data is used to generate features for predictions. This date should be the same or earlier than the end time of the datasets. Use RFC3339 UTC "Zulu" format (for example, 2014-10-02T15:01:23Z).
PREDICTION_PERIODS: The number of consecutive months to produce predictions for, ending with the last full month prior to the prediction end date according to the dataset's timezone.
BQ_OUTPUT_DATASET_NAME: the name of the output BigQuery dataset used for prediction
BQ_OUTPUT_PREDICTION_TABLE: the user-defined identifier for the output BigQuery table used for predictions
BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE: the user-defined identifier for the output explainability BigQuery table used for prediction; remove the optional explainabilityDestination object from the request JSON if you don't want to export to a BigQuery table
WRITE_DISPOSITION: the action that occurs if the destination table already exists; use one of the following values:
- WRITE_EMPTY: Only export data if the BigQuery table is empty.
- WRITE_TRUNCATE: Erase all existing data in the BigQuery table before writing to the table.

Request JSON body:

{
    "model": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/models/MODEL_ID",
    "dataset": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/datasets/DATASET_ID",
    "endTime": "PREDICTION_END_DATE",
    "predictionPeriods": "PREDICTION_PERIODS",
    "outputs": {
      "predictionDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      },
      "explainabilityDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      }
    }
}

To send your request, choose one of these options:

curl

Note: The following command assumes that you have logged in to the gcloud CLI with your user account by running gcloud init or gcloud auth login , or by using Cloud Shell, which automatically logs you into the gcloud CLI . You can check the currently active account by running gcloud auth list.

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

cat > request.json << 'EOF'
{
    "model": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/models/MODEL_ID",
    "dataset": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/datasets/DATASET_ID",
    "endTime": "PREDICTION_END_DATE",
    "predictionPeriods": "PREDICTION_PERIODS",
    "outputs": {
      "predictionDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      },
      "explainabilityDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      }
    }
}
EOF

Then execute the following command to send your REST request:

curl -X POST \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     -H "Content-Type: application/json; charset=utf-8" \
     -d @request.json \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults?prediction_result_id=PREDICTION_RESULTS_ID"

PowerShell

Note: The following command assumes that you have logged in to the gcloud CLI with your user account by running gcloud init or gcloud auth login . You can check the currently active account by running gcloud auth list.

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

@'
{
    "model": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/models/MODEL_ID",
    "dataset": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/datasets/DATASET_ID",
    "endTime": "PREDICTION_END_DATE",
    "predictionPeriods": "PREDICTION_PERIODS",
    "outputs": {
      "predictionDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      },
      "explainabilityDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      }
    }
}
'@  | Out-File -FilePath request.json -Encoding utf8

Then execute the following command to send your REST request:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method POST `
    -Headers $headers `
    -ContentType: "application/json; charset=utf-8" `
    -InFile request.json `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults?prediction_result_id=PREDICTION_RESULTS_ID" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.financialservices.v1.OperationMetadata",
    "createTime": CREATE_TIME,
    "target": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "verb": "create",
    "requestedCancellation": false,
    "apiVersion": "v1"
  },
  "done": false
}

Check for the result

Use the projects.locations.operations.get method to check if prediction results have been created. If the response contains "done": false, repeat the command until the response contains "done": true. These operations can take a few minutes to several hours to complete.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
OPERATION_ID: the identifier for the operation

To send your request, choose one of these options:

curl

Execute the following command:

curl -X GET \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID"

PowerShell

Execute the following command:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method GET `
    -Headers $headers `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.financialservices.v1.OperationMetadata",
    "createTime": CREATE_TIME,
    "endTime": END_TIME,
    "target": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "verb": "create",
    "requestedCancellation": false,
    "apiVersion": "v1"
  },
  "done": true,
  "response": {
    "@type": "type.googleapis.com/dataresidency.monitoring.DataResidencyAugmentedView",
    "tpIds": [
      "i608e8cf4abb2a7d9-tp"
    ]
  }
}

Export metadata

To export metadata from prediction results, use the projects.locations.instances.predictionResults.exportMetadata method.

For more information, see Exported metadata in the AML output data model.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: the user-defined identifier for the instance
PREDICTION_RESULTS_ID: the user-defined identifier for the prediction results
BQ_OUTPUT_DATASET_NAME: a BigQuery dataset in which to export a table that describes the structured metadata of the prediction results
STRUCTURED_METADATA_TABLE: the table to write the structured metadata to
WRITE_DISPOSITION: the action that occurs if the destination table already exists; use one of the following values:
- WRITE_EMPTY: Only export data if the BigQuery table is empty.
- WRITE_TRUNCATE: Erase all existing data in the BigQuery table before writing to the table.

Request JSON body:

{
  "structuredMetadataDestination": {
    "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.STRUCTURED_METADATA_TABLE",
    "writeDisposition": "WRITE_DISPOSITION"
  }
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

cat > request.json << 'EOF'
{
  "structuredMetadataDestination": {
    "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.STRUCTURED_METADATA_TABLE",
    "writeDisposition": "WRITE_DISPOSITION"
  }
}
EOF

Then execute the following command to send your REST request:

curl -X POST \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     -H "Content-Type: application/json; charset=utf-8" \
     -d @request.json \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID:exportMetadata"

PowerShell

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

@'
{
  "structuredMetadataDestination": {
    "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.STRUCTURED_METADATA_TABLE",
    "writeDisposition": "WRITE_DISPOSITION"
  }
}
'@  | Out-File -FilePath request.json -Encoding utf8

Then execute the following command to send your REST request:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method POST `
    -Headers $headers `
    -ContentType: "application/json; charset=utf-8" `
    -InFile request.json `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID:exportMetadata" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.financialservices.v1.OperationMetadata",
    "createTime": CREATE_TIME,
    "target": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "verb": "exportMetadata",
    "requestedCancellation": false,
    "apiVersion": "v1"
  },
  "done": false
}

For more information on how to get the result of the long-running operation (LRO), see Check for the result.

Optional methods

The following prediction results methods are provided as a convenience.

Get prediction results

To get prediction results, use the projects.locations.instances.predictionResults.get method.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: the user-defined identifier for the instance
PREDICTION_RESULTS_ID: the user-defined identifier for the prediction results

To send your request, choose one of these options:

curl

Execute the following command:

curl -X GET \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID"

PowerShell

Execute the following command:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method GET `
    -Headers $headers `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
    "name": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "createTime": CREATE_TIME,
    "updateTime": UPDATE_TIME,
    "state": "ACTIVE",
    "model": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/models/MODEL_ID",
    "dataset": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/datasets/DATASET_ID",
    "endTime": "PREDICTION_END_DATE",
    "predictionPeriods": PREDICTION_PERIODS,
    "outputs": {
      "predictionDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      },
      "explainabilityDestination": {
        "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE",
        "writeDisposition": "WRITE_DISPOSITION"
      }
    },
    "lineOfBusiness": "RETAIL"
}

Update prediction results

To update prediction results, use the projects.locations.instances.predictionResults.patch method.

Not all fields in prediction results can be updated. The following example updates the key-value pair user labels associated with the prediction results.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: the user-defined identifier for the instance
PREDICTION_RESULTS_ID: the user-defined identifier for the prediction results
KEY: The key in a key-value pair used to organize prediction results. See labels for more information.
VALUE: The value in a key-value pair used to organize prediction results. See labels for more information.

Request JSON body:

{
  "labels": {
    "KEY": "VALUE"
  }
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

cat > request.json << 'EOF'
{
  "labels": {
    "KEY": "VALUE"
  }
}
EOF

Then execute the following command to send your REST request:

curl -X PATCH \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     -H "Content-Type: application/json; charset=utf-8" \
     -d @request.json \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID?updateMask=labels"

PowerShell

Save the request body in a file named request.json. Run the following command in the terminal to create or overwrite this file in the current directory:

@'
{
  "labels": {
    "KEY": "VALUE"
  }
}
'@  | Out-File -FilePath request.json -Encoding utf8

Then execute the following command to send your REST request:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method PATCH `
    -Headers $headers `
    -ContentType: "application/json; charset=utf-8" `
    -InFile request.json `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID?updateMask=labels" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.financialservices.v1.OperationMetadata",
    "createTime": CREATE_TIME,
    "target": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "verb": "update",
    "requestedCancellation": false,
    "apiVersion": "v1"
  },
  "done": false
}

For more information on how to get the result of the long-running operation (LRO), see Check for the result.

List the prediction results

To list the prediction results for a given instance, use the projects.locations.instances.predictionResults.list method.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: the user-defined identifier for the instance

To send your request, choose one of these options:

curl

Execute the following command:

curl -X GET \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults"

PowerShell

Execute the following command:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method GET `
    -Headers $headers `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "predictionResults": [
    {
      "name": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
      "createTime": CREATE_TIME,
      "updateTime": UPDATE_TIME,
      "state": "ACTIVE",
      "model": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/models/MODEL_ID",
      "dataset": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/datasets/DATASET_ID",
      "endTime": "PREDICTION_END_DATE",
      "predictionPeriods": PREDICTION_PERIODS,
      "outputs": {
        "predictionDestination": {
          "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_TABLE",
          "writeDisposition": "WRITE_DISPOSITION"
        },
        "explainabilityDestination": {
          "tableUri": "bq://PROJECT_ID.BQ_OUTPUT_DATASET_NAME.BQ_OUTPUT_PREDICTION_EXPLAINABILITY_TABLE",
          "writeDisposition": "WRITE_DISPOSITION"
        }
      },
      "lineOfBusiness": "RETAIL"
    }
  ]
}

Delete prediction results

To delete prediction results, use the projects.locations.instances.predictionResults.delete method.

Before using any of the request data, make the following replacements:

PROJECT_ID: your Google Cloud project ID listed in the IAM Settings
LOCATION: the location of the instance; use one of the supported regions:
- us-central1
- us-east1
- asia-south1
- europe-west1
- europe-west2
- europe-west4
- northamerica-northeast1
- southamerica-east1
INSTANCE_ID: the user-defined identifier for the instance
PREDICTION_RESULTS_ID: the user-defined identifier for the prediction results

To send your request, choose one of these options:

curl

Execute the following command:

curl -X DELETE \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID"

PowerShell

Execute the following command:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method DELETE `
    -Headers $headers `
    -Uri "https://financialservices.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.financialservices.v1.OperationMetadata",
    "createTime": CREATE_TIME,
    "target": "projects/PROJECT_ID/locations/LOCATION/instances/INSTANCE_ID/predictionResults/PREDICTION_RESULTS_ID",
    "verb": "delete",
    "requestedCancellation": false,
    "apiVersion": "v1"
  },
  "done": false
}

For more information on how to get the result of the long-running operation (LRO), see Check for the result.