Get online predictions and explanations

This page shows you how to get online (real-time) predictions and explanations from your tabular classification or regression models using the Google Cloud console or the Vertex AI API.

An online prediction is a synchronous request as opposed to a batch prediction, which is an asynchronous request. Use online predictions when you are making requests in response to application input or in other situations where you require timely inference.

You must deploy a model to an endpoint before that model can be used to serve online predictions. Deploying a model associates physical resources with the model so it can serve online predictions with low latency.

The topics covered are:

  1. Deploy a model to an endpoint
  2. Get an online prediction using your deployed model
  3. Get an online explanation using your deployed model

Before you begin

Before you can get online predictions, you must first train a classification or regression model and evaluate it for accuracy.

Deploy a model to an endpoint

You can deploy more than one model to an endpoint, and you can deploy a model to more than one endpoint. For more information about options and use cases for deploying models, see About deploying models.

Use one of the following methods to deploy a model:

Google Cloud console

  1. In the Google Cloud console, in the Vertex AI section, go to the Models page.

    Go to the Models page

  2. Click the name of the model you want to deploy to open its details page.

  3. Select the Deploy & Test tab.

    If your model is already deployed to any endpoints, they are listed in the Deploy your model section.

  4. Click Deploy to endpoint.

  5. In the Define your endpoint page, configure as follows:

    1. You can choose to deploy your model to a new endpoint or an existing endpoint.

      • To deploy your model to a new endpoint, select Create new endpoint and provide a name for the new endpoint.
      • To deploy your model to an existing endpoint, select Add to existing endpoint and select the endpoint from the drop-down list.
      • You can add more than one model to an endpoint, and you can add a model to more than one endpoint. Learn more.
    2. Click Continue.

  6. In the Model settings page, configure as follows:

    1. If you're deploying your model to a new endpoint, accept 100 for the Traffic split. If you're deploying your model to an existing endpoint that has one or more models deployed to it, you must update the Traffic split percentage for the model you are deploying and the already deployed models so that all of the percentages add up to 100%.

    2. Enter the Minimum number of compute nodes you want to provide for your model.

      This is the number of nodes available to this model at all times. You are charged for the nodes used, whether to handle prediction load or for standby (minimum) nodes, even without prediction traffic. See the pricing page.

    3. Select your Machine type.

      Larger machine resources will increase your prediction performance and increase costs.

    4. Learn how to change the default settings for prediction logging.

    5. Click Continue

  7. In the Model monitoring page, click Continue.

  8. In the Monitoring objectives page, configure as follows:

    1. Enter the location of your training data.
    2. Enter the name of the target column.
  9. Click Deploy to deploy your model to the endpoint.

API

When you deploy a model using the Vertex AI API, you complete the following steps:

  1. Create an endpoint if needed.
  2. Get the endpoint ID.
  3. Deploy the model to the endpoint.

Create an endpoint

If you are deploying a model to an existing endpoint, you can skip this step.

gcloud

The following example uses the gcloud ai endpoints create command:

  gcloud ai endpoints create \
    --region=LOCATION \
    --display-name=ENDPOINT_NAME

Replace the following:

  • LOCATION_ID: The region where you are using Vertex AI.
  • ENDPOINT_NAME: The display name for the endpoint.

    The Google Cloud CLI tool might take a few seconds to create the endpoint.

REST

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

  • LOCATION_ID: Your region.
  • PROJECT_ID: Your project ID.
  • ENDPOINT_NAME: The display name for the endpoint.

HTTP method and URL:

POST https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints

Request JSON body:

{
  "display_name": "ENDPOINT_NAME"
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_NUMBER/locations/LOCATION_ID/endpoints/ENDPOINT_ID/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.aiplatform.v1.CreateEndpointOperationMetadata",
    "genericMetadata": {
      "createTime": "2020-11-05T17:45:42.812656Z",
      "updateTime": "2020-11-05T17:45:42.812656Z"
    }
  }
}
You can poll for the status of the operation until the response includes "done": true.

Java

Before trying this sample, follow the Java setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Java API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.


import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.aiplatform.v1.CreateEndpointOperationMetadata;
import com.google.cloud.aiplatform.v1.Endpoint;
import com.google.cloud.aiplatform.v1.EndpointServiceClient;
import com.google.cloud.aiplatform.v1.EndpointServiceSettings;
import com.google.cloud.aiplatform.v1.LocationName;
import java.io.IOException;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class CreateEndpointSample {

  public static void main(String[] args)
      throws IOException, InterruptedException, ExecutionException, TimeoutException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "YOUR_PROJECT_ID";
    String endpointDisplayName = "YOUR_ENDPOINT_DISPLAY_NAME";
    createEndpointSample(project, endpointDisplayName);
  }

  static void createEndpointSample(String project, String endpointDisplayName)
      throws IOException, InterruptedException, ExecutionException, TimeoutException {
    EndpointServiceSettings endpointServiceSettings =
        EndpointServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (EndpointServiceClient endpointServiceClient =
        EndpointServiceClient.create(endpointServiceSettings)) {
      String location = "us-central1";
      LocationName locationName = LocationName.of(project, location);
      Endpoint endpoint = Endpoint.newBuilder().setDisplayName(endpointDisplayName).build();

      OperationFuture<Endpoint, CreateEndpointOperationMetadata> endpointFuture =
          endpointServiceClient.createEndpointAsync(locationName, endpoint);
      System.out.format("Operation name: %s\n", endpointFuture.getInitialFuture().get().getName());
      System.out.println("Waiting for operation to finish...");
      Endpoint endpointResponse = endpointFuture.get(300, TimeUnit.SECONDS);

      System.out.println("Create Endpoint Response");
      System.out.format("Name: %s\n", endpointResponse.getName());
      System.out.format("Display Name: %s\n", endpointResponse.getDisplayName());
      System.out.format("Description: %s\n", endpointResponse.getDescription());
      System.out.format("Labels: %s\n", endpointResponse.getLabelsMap());
      System.out.format("Create Time: %s\n", endpointResponse.getCreateTime());
      System.out.format("Update Time: %s\n", endpointResponse.getUpdateTime());
    }
  }
}

Node.js

Before trying this sample, follow the Node.js setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Node.js API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */

// const endpointDisplayName = 'YOUR_ENDPOINT_DISPLAY_NAME';
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';

// Imports the Google Cloud Endpoint Service Client library
const {EndpointServiceClient} = require('@google-cloud/aiplatform');

// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};

// Instantiates a client
const endpointServiceClient = new EndpointServiceClient(clientOptions);

async function createEndpoint() {
  // Configure the parent resource
  const parent = `projects/${project}/locations/${location}`;
  const endpoint = {
    displayName: endpointDisplayName,
  };
  const request = {
    parent,
    endpoint,
  };

  // Get and print out a list of all the endpoints for this resource
  const [response] = await endpointServiceClient.createEndpoint(request);
  console.log(`Long running operation : ${response.name}`);

  // Wait for operation to complete
  await response.promise();
  const result = response.result;

  console.log('Create endpoint response');
  console.log(`\tName : ${result.name}`);
  console.log(`\tDisplay name : ${result.displayName}`);
  console.log(`\tDescription : ${result.description}`);
  console.log(`\tLabels : ${JSON.stringify(result.labels)}`);
  console.log(`\tCreate time : ${JSON.stringify(result.createTime)}`);
  console.log(`\tUpdate time : ${JSON.stringify(result.updateTime)}`);
}
createEndpoint();

Python

To learn how to install or update the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python. For more information, see the Python API reference documentation.

def create_endpoint_sample(
    project: str,
    display_name: str,
    location: str,
):
    aiplatform.init(project=project, location=location)

    endpoint = aiplatform.Endpoint.create(
        display_name=display_name,
        project=project,
        location=location,
    )

    print(endpoint.display_name)
    print(endpoint.resource_name)
    return endpoint

Get the endpoint ID

You need the endpoint ID to deploy the model.

gcloud

The following example uses the gcloud ai endpoints list command:

  gcloud ai endpoints list \
    --region=LOCATION \
    --filter=display_name=ENDPOINT_NAME

Replace the following:

  • LOCATION_ID: The region where you are using Vertex AI.
  • ENDPOINT_NAME: The display name for the endpoint.

    Note the number that appears in the ENDPOINT_ID column. Use this ID in the following step.

REST

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

  • LOCATION_ID: The region where you are using Vertex AI.
  • PROJECT_ID: Your project ID.
  • ENDPOINT_NAME: The display name for the endpoint.

HTTP method and URL:

GET https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints?filter=display_name=ENDPOINT_NAME

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

{
  "endpoints": [
    {
      "name": "projects/PROJECT_NUMBER/locations/LOCATION_ID/endpoints/ENDPOINT_ID",
      "displayName": "ENDPOINT_NAME",
      "etag": "AMEw9yPz5pf4PwBHbRWOGh0PcAxUdjbdX2Jm3QO_amguy3DbZGP5Oi_YUKRywIE-BtLx",
      "createTime": "2020-04-17T18:31:11.585169Z",
      "updateTime": "2020-04-17T18:35:08.568959Z"
    }
  ]
}
Note the ENDPOINT_ID.

Deploy the model

Select the tab below for your language or environment:

gcloud

The following examples use the gcloud ai endpoints deploy-model command.

The following example deploys a Model to an Endpoint without using GPUs to accelerate prediction serving and without splitting traffic between multiple DeployedModel resources:

Before using any of the command data below, make the following replacements:

  • ENDPOINT_ID: The ID for the endpoint.
  • LOCATION_ID: The region where you are using Vertex AI.
  • MODEL_ID: The ID for the model to be deployed.
  • DEPLOYED_MODEL_NAME: A name for the DeployedModel. You can use the display name of the Model for the DeployedModel as well.
  • MACHINE_TYPE: Optional. The machine resources used for each node of this deployment. Its default setting is n1-standard-2. Learn more about machine types.
  • MIN_REPLICA_COUNT: The minimum number of nodes for this deployment. The node count can be increased or decreased as required by the prediction load, up to the maximum number of nodes and never fewer than this number of nodes. This value must be greater than or equal to 1. If the --min-replica-count flag is omitted, the value defaults to 1.
  • MAX_REPLICA_COUNT: The maximum number of nodes for this deployment. The node count can be increased or decreased as required by the prediction load, up to this number of nodes and never fewer than the minimum number of nodes. If you omit the --max-replica-count flag, then maximum number of nodes is set to the value of --min-replica-count.

Execute the gcloud ai endpoints deploy-model command:

Linux, macOS, or Cloud Shell

gcloud ai endpoints deploy-model ENDPOINT_ID\
  --region=LOCATION_ID \
  --model=MODEL_ID \
  --display-name=DEPLOYED_MODEL_NAME \
  --machine-type=MACHINE_TYPE \
  --min-replica-count=MIN_REPLICA_COUNT \
  --max-replica-count=MAX_REPLICA_COUNT \
  --traffic-split=0=100

Windows (PowerShell)

gcloud ai endpoints deploy-model ENDPOINT_ID`
  --region=LOCATION_ID `
  --model=MODEL_ID `
  --display-name=DEPLOYED_MODEL_NAME `
  --machine-type=MACHINE_TYPE `
  --min-replica-count=MIN_REPLICA_COUNT `
  --max-replica-count=MAX_REPLICA_COUNT `
  --traffic-split=0=100

Windows (cmd.exe)

gcloud ai endpoints deploy-model ENDPOINT_ID^
  --region=LOCATION_ID ^
  --model=MODEL_ID ^
  --display-name=DEPLOYED_MODEL_NAME ^
  --machine-type=MACHINE_TYPE ^
  --min-replica-count=MIN_REPLICA_COUNT ^
  --max-replica-count=MAX_REPLICA_COUNT ^
  --traffic-split=0=100
 

Splitting traffic

The --traffic-split=0=100 flag in the preceding examples sends 100% of prediction traffic that the Endpoint receives to the new DeployedModel, which is represented by the temporary ID 0. If your Endpoint already has other DeployedModel resources, then you can split traffic between the new DeployedModel and the old ones. For example, to send 20% of traffic to the new DeployedModel and 80% to an older one, run the following command.

Before using any of the command data below, make the following replacements:

  • OLD_DEPLOYED_MODEL_ID: the ID of the existing DeployedModel.

Execute the gcloud ai endpoints deploy-model command:

Linux, macOS, or Cloud Shell

gcloud ai endpoints deploy-model ENDPOINT_ID\
  --region=LOCATION_ID \
  --model=MODEL_ID \
  --display-name=DEPLOYED_MODEL_NAME \ 
  --machine-type=MACHINE_TYPE \
  --min-replica-count=MIN_REPLICA_COUNT \
  --max-replica-count=MAX_REPLICA_COUNT \
  --traffic-split=0=20,OLD_DEPLOYED_MODEL_ID=80

Windows (PowerShell)

gcloud ai endpoints deploy-model ENDPOINT_ID`
  --region=LOCATION_ID `
  --model=MODEL_ID `
  --display-name=DEPLOYED_MODEL_NAME \ 
  --machine-type=MACHINE_TYPE `
  --min-replica-count=MIN_REPLICA_COUNT `
  --max-replica-count=MAX_REPLICA_COUNT `
  --traffic-split=0=20,OLD_DEPLOYED_MODEL_ID=80

Windows (cmd.exe)

gcloud ai endpoints deploy-model ENDPOINT_ID^
  --region=LOCATION_ID ^
  --model=MODEL_ID ^
  --display-name=DEPLOYED_MODEL_NAME \ 
  --machine-type=MACHINE_TYPE ^
  --min-replica-count=MIN_REPLICA_COUNT ^
  --max-replica-count=MAX_REPLICA_COUNT ^
  --traffic-split=0=20,OLD_DEPLOYED_MODEL_ID=80
 

REST

You use the endpoints.predict method to request an online prediction.

Deploy the model.

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

  • LOCATION_ID: The region where you are using Vertex AI.
  • PROJECT_ID: Your project ID.
  • ENDPOINT_ID: The ID for the endpoint.
  • MODEL_ID: The ID for the model to be deployed.
  • DEPLOYED_MODEL_NAME: A name for the DeployedModel. You can use the display name of the Model for the DeployedModel as well.
  • MACHINE_TYPE: Optional. The machine resources used for each node of this deployment. Its default setting is n1-standard-2. Learn more about machine types.
  • ACCELERATOR_TYPE: The type of accelerator to be attached to the machine. Optional if ACCELERATOR_COUNT is not specified or is zero. Not recommended for AutoML models or custom-trained models that are using non-GPU images. Learn more.
  • ACCELERATOR_COUNT: The number of accelerators for each replica to use. Optional. Should be zero or unspecified for AutoML models or custom-trained models that are using non-GPU images.
  • MIN_REPLICA_COUNT: The minimum number of nodes for this deployment. The node count can be increased or decreased as required by the prediction load, up to the maximum number of nodes and never fewer than this number of nodes. This value must be greater than or equal to 1.
  • MAX_REPLICA_COUNT: The maximum number of nodes for this deployment. The node count can be increased or decreased as required by the prediction load, up to this number of nodes and never fewer than the minimum number of nodes.
  • TRAFFIC_SPLIT_THIS_MODEL: The percentage of the prediction traffic to this endpoint to be routed to the model being deployed with this operation. Defaults to 100. All traffic percentages must add up to 100. Learn more about traffic splits.
  • DEPLOYED_MODEL_ID_N: Optional. If other models are deployed to this endpoint, you must update their traffic split percentages so that all percentages add up to 100.
  • TRAFFIC_SPLIT_MODEL_N: The traffic split percentage value for the deployed model id key.
  • PROJECT_NUMBER: Your project's automatically generated project number

HTTP method and URL:

POST https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:deployModel

Request JSON body:

{
  "deployedModel": {
    "model": "projects/PROJECT/locations/us-central1/models/MODEL_ID",
    "displayName": "DEPLOYED_MODEL_NAME",
    "dedicatedResources": {
       "machineSpec": {
         "machineType": "MACHINE_TYPE",
         "acceleratorType": "ACCELERATOR_TYPE",
         "acceleratorCount": "ACCELERATOR_COUNT"
       },
       "minReplicaCount": MIN_REPLICA_COUNT,
       "maxReplicaCount": MAX_REPLICA_COUNT
     },
  },
  "trafficSplit": {
    "0": TRAFFIC_SPLIT_THIS_MODEL,
    "DEPLOYED_MODEL_ID_1": TRAFFIC_SPLIT_MODEL_1,
    "DEPLOYED_MODEL_ID_2": TRAFFIC_SPLIT_MODEL_2
  },
}

To send your request, expand one of these options:

You should receive a JSON response similar to the following:

{
  "name": "projects/PROJECT_ID/locations/LOCATION/endpoints/ENDPOINT_ID/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.aiplatform.v1.DeployModelOperationMetadata",
    "genericMetadata": {
      "createTime": "2020-10-19T17:53:16.502088Z",
      "updateTime": "2020-10-19T17:53:16.502088Z"
    }
  }
}

Java

Before trying this sample, follow the Java setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Java API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.aiplatform.v1.DedicatedResources;
import com.google.cloud.aiplatform.v1.DeployModelOperationMetadata;
import com.google.cloud.aiplatform.v1.DeployModelResponse;
import com.google.cloud.aiplatform.v1.DeployedModel;
import com.google.cloud.aiplatform.v1.EndpointName;
import com.google.cloud.aiplatform.v1.EndpointServiceClient;
import com.google.cloud.aiplatform.v1.EndpointServiceSettings;
import com.google.cloud.aiplatform.v1.MachineSpec;
import com.google.cloud.aiplatform.v1.ModelName;
import java.io.IOException;
import java.util.HashMap;
import java.util.Map;
import java.util.concurrent.ExecutionException;

public class DeployModelCustomTrainedModelSample {

  public static void main(String[] args)
      throws IOException, ExecutionException, InterruptedException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "PROJECT";
    String endpointId = "ENDPOINT_ID";
    String modelName = "MODEL_NAME";
    String deployedModelDisplayName = "DEPLOYED_MODEL_DISPLAY_NAME";
    deployModelCustomTrainedModelSample(project, endpointId, modelName, deployedModelDisplayName);
  }

  static void deployModelCustomTrainedModelSample(
      String project, String endpointId, String model, String deployedModelDisplayName)
      throws IOException, ExecutionException, InterruptedException {
    EndpointServiceSettings settings =
        EndpointServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();
    String location = "us-central1";

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (EndpointServiceClient client = EndpointServiceClient.create(settings)) {
      MachineSpec machineSpec = MachineSpec.newBuilder().setMachineType("n1-standard-2").build();
      DedicatedResources dedicatedResources =
          DedicatedResources.newBuilder().setMinReplicaCount(1).setMachineSpec(machineSpec).build();

      String modelName = ModelName.of(project, location, model).toString();
      DeployedModel deployedModel =
          DeployedModel.newBuilder()
              .setModel(modelName)
              .setDisplayName(deployedModelDisplayName)
              // `dedicated_resources` must be used for non-AutoML models
              .setDedicatedResources(dedicatedResources)
              .build();
      // key '0' assigns traffic for the newly deployed model
      // Traffic percentage values must add up to 100
      // Leave dictionary empty if endpoint should not accept any traffic
      Map<String, Integer> trafficSplit = new HashMap<>();
      trafficSplit.put("0", 100);
      EndpointName endpoint = EndpointName.of(project, location, endpointId);
      OperationFuture<DeployModelResponse, DeployModelOperationMetadata> response =
          client.deployModelAsync(endpoint, deployedModel, trafficSplit);

      // You can use OperationFuture.getInitialFuture to get a future representing the initial
      // response to the request, which contains information while the operation is in progress.
      System.out.format("Operation name: %s\n", response.getInitialFuture().get().getName());

      // OperationFuture.get() will block until the operation is finished.
      DeployModelResponse deployModelResponse = response.get();
      System.out.format("deployModelResponse: %s\n", deployModelResponse);
    }
  }
}

Python

To learn how to install or update the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python. For more information, see the Python API reference documentation.

def deploy_model_with_dedicated_resources_sample(
    project,
    location,
    model_name: str,
    machine_type: str,
    endpoint: Optional[aiplatform.Endpoint] = None,
    deployed_model_display_name: Optional[str] = None,
    traffic_percentage: Optional[int] = 0,
    traffic_split: Optional[Dict[str, int]] = None,
    min_replica_count: int = 1,
    max_replica_count: int = 1,
    accelerator_type: Optional[str] = None,
    accelerator_count: Optional[int] = None,
    explanation_metadata: Optional[explain.ExplanationMetadata] = None,
    explanation_parameters: Optional[explain.ExplanationParameters] = None,
    metadata: Optional[Sequence[Tuple[str, str]]] = (),
    sync: bool = True,
):
    """
    model_name: A fully-qualified model resource name or model ID.
          Example: "projects/123/locations/us-central1/models/456" or
          "456" when project and location are initialized or passed.
    """

    aiplatform.init(project=project, location=location)

    model = aiplatform.Model(model_name=model_name)

    # The explanation_metadata and explanation_parameters should only be
    # provided for a custom trained model and not an AutoML model.
    model.deploy(
        endpoint=endpoint,
        deployed_model_display_name=deployed_model_display_name,
        traffic_percentage=traffic_percentage,
        traffic_split=traffic_split,
        machine_type=machine_type,
        min_replica_count=min_replica_count,
        max_replica_count=max_replica_count,
        accelerator_type=accelerator_type,
        accelerator_count=accelerator_count,
        explanation_metadata=explanation_metadata,
        explanation_parameters=explanation_parameters,
        metadata=metadata,
        sync=sync,
    )

    model.wait()

    print(model.display_name)
    print(model.resource_name)
    return model

Node.js

Before trying this sample, follow the Node.js setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Node.js API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

const automl = require('@google-cloud/automl');
const client = new automl.v1beta1.AutoMlClient();

/**
 * Demonstrates using the AutoML client to create a model.
 * TODO(developer): Uncomment the following lines before running the sample.
 */
// const projectId = '[PROJECT_ID]' e.g., "my-gcloud-project";
// const computeRegion = '[REGION_NAME]' e.g., "us-central1";
// const datasetId = '[DATASET_ID]' e.g., "TBL2246891593778855936";
// const tableId = '[TABLE_ID]' e.g., "1991013247762825216";
// const columnId = '[COLUMN_ID]' e.g., "773141392279994368";
// const modelName = '[MODEL_NAME]' e.g., "testModel";
// const trainBudget = '[TRAIN_BUDGET]' e.g., "1000",
// `Train budget in milli node hours`;

// A resource that represents Google Cloud Platform location.
const projectLocation = client.locationPath(projectId, computeRegion);

// Get the full path of the column.
const columnSpecId = client.columnSpecPath(
  projectId,
  computeRegion,
  datasetId,
  tableId,
  columnId
);

// Set target column to train the model.
const targetColumnSpec = {name: columnSpecId};

// Set tables model metadata.
const tablesModelMetadata = {
  targetColumnSpec: targetColumnSpec,
  trainBudgetMilliNodeHours: trainBudget,
};

// Set datasetId, model name and model metadata for the dataset.
const myModel = {
  datasetId: datasetId,
  displayName: modelName,
  tablesModelMetadata: tablesModelMetadata,
};

// Create a model with the model metadata in the region.
client
  .createModel({parent: projectLocation, model: myModel})
  .then(responses => {
    const initialApiResponse = responses[1];
    console.log(`Training operation name: ${initialApiResponse.name}`);
    console.log('Training started...');
  })
  .catch(err => {
    console.error(err);
  });

Learn how to change the default settings for prediction logging.

Get operation status

Some requests start long-running operations that require time to complete. These requests return an operation name, which you can use to view the operation's status or cancel the operation. Vertex AI provides helper methods to make calls against long-running operations. For more information, see Working with long-running operations.

Get an online prediction using your deployed model

To make an online prediction, submit one or more test items to a model for analysis, and the model returns results that are based on your model's objective. Use the Google Cloud console or the Vertex AI API to request an online prediction.

Google Cloud console

  1. In the Google Cloud console, in the Vertex AI section, go to the Models page.

    Go to the Models page

  2. From the list of models, click the name of the model to request predictions from.

  3. Select the Deploy & test tab.

  4. Under the Test your model section, add test items to request a prediction. The baseline prediction data is filled in for you, or you can enter your own prediction data and click Predict.

    After the prediction is complete, Vertex AI returns the results in the console.

API: Classification

gcloud

  1. Create a file named request.json with the following contents:

          {
      "instances": [
        {
          PREDICTION_DATA_ROW
        }
      ]
    }
        

    Replace the following:

    • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of strings, and a category, the row of data might look like the following example request:

      "length":3.6,
      "material":"cotton",
      "tag_array": ["abc","def"]
      

      A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  2. Run the following command:

    gcloud ai endpoints predict ENDPOINT_ID \
      --region=LOCATION_ID \
      --json-request=request.json

    Replace the following:

    • ENDPOINT_ID: The ID for the endpoint.
    • LOCATION_ID: The region where you are using Vertex AI.

REST

You use the endpoints.predict method to request an online prediction.

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

  • LOCATION_ID: Region where Endpoint is located. For example, us-central1.
  • PROJECT_ID: Your project ID.
  • ENDPOINT_ID: The ID for the endpoint.
  • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of strings, and a category, the row of data might look like the following example request:

    "length":3.6,
    "material":"cotton",
    "tag_array": ["abc","def"]
    

    A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  • DEPLOYED_MODEL_ID: Output by the predict method, and accepted as input by the explain method. The ID of the model used to generate the prediction. If you need to request explanations for a previously requested prediction, and you have more than one model deployed, you can use this ID to ensure that the explanations are returned for the same model that provided the previous prediction.

HTTP method and URL:

POST https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict

Request JSON body:

{
  "instances": [
    {
      PREDICTION_DATA_ROW
    }
  ]
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json, and execute the following command:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict"

PowerShell

Save the request body in a file named request.json, and execute the following command:

$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://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict" | Select-Object -Expand Content

You should receive a JSON response similar to the following:

   {
     "predictions": [
      {
         "scores": [
           0.96771615743637085,
           0.032283786684274673
         ],
         "classes": [
           "0",
           "1"
         ]
      }
     ]
     "deployedModelId": "2429510197"
   }
   

Java

Before trying this sample, follow the Java setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Java API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.


import com.google.cloud.aiplatform.util.ValueConverter;
import com.google.cloud.aiplatform.v1.EndpointName;
import com.google.cloud.aiplatform.v1.PredictResponse;
import com.google.cloud.aiplatform.v1.PredictionServiceClient;
import com.google.cloud.aiplatform.v1.PredictionServiceSettings;
import com.google.cloud.aiplatform.v1.schema.predict.prediction.TabularClassificationPredictionResult;
import com.google.protobuf.ListValue;
import com.google.protobuf.Value;
import com.google.protobuf.util.JsonFormat;
import java.io.IOException;
import java.util.List;

public class PredictTabularClassificationSample {

  public static void main(String[] args) throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "YOUR_PROJECT_ID";
    String instance = "[{ “feature_column_a”: “value”, “feature_column_b”: “value”}]";
    String endpointId = "YOUR_ENDPOINT_ID";
    predictTabularClassification(instance, project, endpointId);
  }

  static void predictTabularClassification(String instance, String project, String endpointId)
      throws IOException {
    PredictionServiceSettings predictionServiceSettings =
        PredictionServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (PredictionServiceClient predictionServiceClient =
        PredictionServiceClient.create(predictionServiceSettings)) {
      String location = "us-central1";
      EndpointName endpointName = EndpointName.of(project, location, endpointId);

      ListValue.Builder listValue = ListValue.newBuilder();
      JsonFormat.parser().merge(instance, listValue);
      List<Value> instanceList = listValue.getValuesList();

      Value parameters = Value.newBuilder().setListValue(listValue).build();
      PredictResponse predictResponse =
          predictionServiceClient.predict(endpointName, instanceList, parameters);
      System.out.println("Predict Tabular Classification Response");
      System.out.format("\tDeployed Model Id: %s\n", predictResponse.getDeployedModelId());

      System.out.println("Predictions");
      for (Value prediction : predictResponse.getPredictionsList()) {
        TabularClassificationPredictionResult.Builder resultBuilder =
            TabularClassificationPredictionResult.newBuilder();
        TabularClassificationPredictionResult result =
            (TabularClassificationPredictionResult)
                ValueConverter.fromValue(resultBuilder, prediction);

        for (int i = 0; i < result.getClassesCount(); i++) {
          System.out.printf("\tClass: %s", result.getClasses(i));
          System.out.printf("\tScore: %f", result.getScores(i));
        }
      }
    }
  }
}

Node.js

Before trying this sample, follow the Node.js setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Node.js API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */

// const endpointId = 'YOUR_ENDPOINT_ID';
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';
const aiplatform = require('@google-cloud/aiplatform');
const {prediction} =
  aiplatform.protos.google.cloud.aiplatform.v1.schema.predict;

// Imports the Google Cloud Prediction service client
const {PredictionServiceClient} = aiplatform.v1;

// Import the helper module for converting arbitrary protobuf.Value objects.
const {helpers} = aiplatform;

// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};

// Instantiates a client
const predictionServiceClient = new PredictionServiceClient(clientOptions);

async function predictTablesClassification() {
  // Configure the endpoint resource
  const endpoint = `projects/${project}/locations/${location}/endpoints/${endpointId}`;
  const parameters = helpers.toValue({});

  const instance = helpers.toValue({
    petal_length: '1.4',
    petal_width: '1.3',
    sepal_length: '5.1',
    sepal_width: '2.8',
  });

  const instances = [instance];
  const request = {
    endpoint,
    instances,
    parameters,
  };

  // Predict request
  const [response] = await predictionServiceClient.predict(request);

  console.log('Predict tabular classification response');
  console.log(`\tDeployed model id : ${response.deployedModelId}\n`);
  const predictions = response.predictions;
  console.log('Predictions :');
  for (const predictionResultVal of predictions) {
    const predictionResultObj =
      prediction.TabularClassificationPredictionResult.fromValue(
        predictionResultVal
      );
    for (const [i, class_] of predictionResultObj.classes.entries()) {
      console.log(`\tClass: ${class_}`);
      console.log(`\tScore: ${predictionResultObj.scores[i]}\n\n`);
    }
  }
}
predictTablesClassification();

Python

To learn how to install or update the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python. For more information, see the Python API reference documentation.

def predict_tabular_classification_sample(
    project: str,
    location: str,
    endpoint_name: str,
    instances: List[Dict],
):
    """
    Args
        project: Your project ID or project number.
        location: Region where Endpoint is located. For example, 'us-central1'.
        endpoint_name: A fully qualified endpoint name or endpoint ID. Example: "projects/123/locations/us-central1/endpoints/456" or
               "456" when project and location are initialized or passed.
        instances: A list of one or more instances (examples) to return a prediction for.
    """
    aiplatform.init(project=project, location=location)

    endpoint = aiplatform.Endpoint(endpoint_name)

    response = endpoint.predict(instances=instances)

    for prediction_ in response.predictions:
        print(prediction_)

API: Regression

gcloud

  1. Create a file named `request.json` with the following contents:

          {
      "instances": [
        {
          PREDICTION_DATA_ROW
        }
      ]
    }
        

    Replace the following:

    • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of numbers, and a category, the row of data might look like the following example request:

      "age":3.6,
      "sq_ft":5392,
      "code": "90331"
      

      A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  2. Run the following command:

    gcloud ai endpoints predict ENDPOINT_ID \
      --region=LOCATION_ID \
      --json-request=request.json

    Replace the following:

    • ENDPOINT_ID: The ID for the endpoint.
    • LOCATION_ID: The region where you are using Vertex AI.

REST

You use the endpoints.predict method to request an online prediction.

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

  • LOCATION_ID: Region where Endpoint is located. For example, us-central1.
  • PROJECT_ID: Your project ID.
  • ENDPOINT_ID: The ID for the endpoint.
  • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of numbers, and a category, the row of data might look like the following example request:

    "age":3.6,
    "sq_ft":5392,
    "code": "90331"
    

    A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  • DEPLOYED_MODEL_ID: Output by the predict method, and accepted as input by the explain method. The ID of the model used to generate the prediction. If you need to request explanations for a previously requested prediction, and you have more than one model deployed, you can use this ID to ensure that the explanations are returned for the same model that provided the previous prediction.

HTTP method and URL:

POST https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict

Request JSON body:

{
  "instances": [
    {
      PREDICTION_DATA_ROW
    }
  ]
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json, and execute the following command:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict"

PowerShell

Save the request body in a file named request.json, and execute the following command:

$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://LOCATION_ID-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION_ID/endpoints/ENDPOINT_ID:predict" | Select-Object -Expand Content

You should receive a JSON response similar to the following:


{
  "predictions": [
    [
      {
        "value": 65.14233,
        "lower_bound": 4.6572,
        "upper_bound": 164.0279
      }
    ]
  ],
  "deployedModelId": "DEPLOYED_MODEL_ID"
}

Java

Before trying this sample, follow the Java setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Java API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.


import com.google.cloud.aiplatform.util.ValueConverter;
import com.google.cloud.aiplatform.v1.EndpointName;
import com.google.cloud.aiplatform.v1.PredictResponse;
import com.google.cloud.aiplatform.v1.PredictionServiceClient;
import com.google.cloud.aiplatform.v1.PredictionServiceSettings;
import com.google.cloud.aiplatform.v1.schema.predict.prediction.TabularRegressionPredictionResult;
import com.google.protobuf.ListValue;
import com.google.protobuf.Value;
import com.google.protobuf.util.JsonFormat;
import java.io.IOException;
import java.util.List;

public class PredictTabularRegressionSample {

  public static void main(String[] args) throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "YOUR_PROJECT_ID";
    String instance = "[{ “feature_column_a”: “value”, “feature_column_b”: “value”}]";
    String endpointId = "YOUR_ENDPOINT_ID";
    predictTabularRegression(instance, project, endpointId);
  }

  static void predictTabularRegression(String instance, String project, String endpointId)
      throws IOException {
    PredictionServiceSettings predictionServiceSettings =
        PredictionServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (PredictionServiceClient predictionServiceClient =
        PredictionServiceClient.create(predictionServiceSettings)) {
      String location = "us-central1";
      EndpointName endpointName = EndpointName.of(project, location, endpointId);

      ListValue.Builder listValue = ListValue.newBuilder();
      JsonFormat.parser().merge(instance, listValue);
      List<Value> instanceList = listValue.getValuesList();

      Value parameters = Value.newBuilder().setListValue(listValue).build();
      PredictResponse predictResponse =
          predictionServiceClient.predict(endpointName, instanceList, parameters);
      System.out.println("Predict Tabular Regression Response");
      System.out.format("\tDisplay Model Id: %s\n", predictResponse.getDeployedModelId());

      System.out.println("Predictions");
      for (Value prediction : predictResponse.getPredictionsList()) {
        TabularRegressionPredictionResult.Builder resultBuilder =
            TabularRegressionPredictionResult.newBuilder();

        TabularRegressionPredictionResult result =
            (TabularRegressionPredictionResult) ValueConverter.fromValue(resultBuilder, prediction);

        System.out.printf("\tUpper bound: %f\n", result.getUpperBound());
        System.out.printf("\tLower bound: %f\n", result.getLowerBound());
        System.out.printf("\tValue: %f\n", result.getValue());
      }
    }
  }
}

Node.js

Before trying this sample, follow the Node.js setup instructions in the Vertex AI quickstart using client libraries. For more information, see the Vertex AI Node.js API reference documentation.

To authenticate to Vertex AI, set up Application Default Credentials. For more information, see Set up authentication for a local development environment.

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */

// const endpointId = 'YOUR_ENDPOINT_ID';
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';
const aiplatform = require('@google-cloud/aiplatform');
const {prediction} =
  aiplatform.protos.google.cloud.aiplatform.v1.schema.predict;

// Imports the Google Cloud Prediction service client
const {PredictionServiceClient} = aiplatform.v1;

// Import the helper module for converting arbitrary protobuf.Value objects.
const {helpers} = aiplatform;

// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};

// Instantiates a client
const predictionServiceClient = new PredictionServiceClient(clientOptions);

async function predictTablesRegression() {
  // Configure the endpoint resource
  const endpoint = `projects/${project}/locations/${location}/endpoints/${endpointId}`;
  const parameters = helpers.toValue({});

  // TODO (erschmid): Make this less painful
  const instance = helpers.toValue({
    BOOLEAN_2unique_NULLABLE: false,
    DATETIME_1unique_NULLABLE: '2019-01-01 00:00:00',
    DATE_1unique_NULLABLE: '2019-01-01',
    FLOAT_5000unique_NULLABLE: 1611,
    FLOAT_5000unique_REPEATED: [2320, 1192],
    INTEGER_5000unique_NULLABLE: '8',
    NUMERIC_5000unique_NULLABLE: 16,
    STRING_5000unique_NULLABLE: 'str-2',
    STRUCT_NULLABLE: {
      BOOLEAN_2unique_NULLABLE: false,
      DATE_1unique_NULLABLE: '2019-01-01',
      DATETIME_1unique_NULLABLE: '2019-01-01 00:00:00',
      FLOAT_5000unique_NULLABLE: 1308,
      FLOAT_5000unique_REPEATED: [2323, 1178],
      FLOAT_5000unique_REQUIRED: 3089,
      INTEGER_5000unique_NULLABLE: '1777',
      NUMERIC_5000unique_NULLABLE: 3323,
      TIME_1unique_NULLABLE: '23:59:59.999999',
      STRING_5000unique_NULLABLE: 'str-49',
      TIMESTAMP_1unique_NULLABLE: '1546387199999999',
    },
    TIMESTAMP_1unique_NULLABLE: '1546387199999999',
    TIME_1unique_NULLABLE: '23:59:59.999999',
  });

  const instances = [instance];
  const request = {
    endpoint,
    instances,
    parameters,
  };

  // Predict request
  const [response] = await predictionServiceClient.predict(request);

  console.log('Predict tabular regression response');
  console.log(`\tDeployed model id : ${response.deployedModelId}`);
  const predictions = response.predictions;
  console.log('\tPredictions :');
  for (const predictionResultVal of predictions) {
    const predictionResultObj =
      prediction.TabularRegressionPredictionResult.fromValue(
        predictionResultVal
      );
    console.log(`\tUpper bound: ${predictionResultObj.upper_bound}`);
    console.log(`\tLower bound: ${predictionResultObj.lower_bound}`);
    console.log(`\tLower bound: ${predictionResultObj.value}`);
  }
}
predictTablesRegression();

Python

To learn how to install or update the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python. For more information, see the Python API reference documentation.

def predict_tabular_regression_sample(
    project: str,
    location: str,
    endpoint_name: str,
    instances: List[Dict],
):
    aiplatform.init(project=project, location=location)

    endpoint = aiplatform.Endpoint(endpoint_name)

    response = endpoint.predict(instances=instances)

    for prediction_ in response.predictions:
        print(prediction_)

Interpret prediction results

Classification

Classification models return a confidence score.

The confidence score communicates how strongly your model associates each class or label with a test item. The higher the number, the higher the model's confidence that the label should be applied to that item. You decide how high the confidence score must be for you to accept the model's results.

Regression

Regression models return a prediction value. For BigQuery destinations, they also return a prediction interval. The prediction interval provides a range of values that the model has 95% confidence contain the actual result.

Get an online explanation using your deployed model

You can request a prediction with explanations (also called feature attributions) to see how your model arrived at a prediction. The local feature importance values tell you how much each feature contributed to the prediction result. Feature attributions are included in Vertex AI predictions through Vertex Explainable AI.

Console

When you use the Google Cloud console to request an online prediction, the local feature importance values are automatically returned.

If you used the pre-filled prediction values, the local feature importance values are all zero. This is because the pre-filled values are the baseline prediction data, so the prediction returned is the baseline prediction value.

gcloud

  1. Create a file named request.json with the following contents:

    {
      "instances": [
        {
          PREDICTION_DATA_ROW
        }
      ]
    }
    

    Replace the following:

    • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of strings, and a category, the row of data might look like the following example request:

      "length":3.6,
      "material":"cotton",
      "tag_array": ["abc","def"]
      

      A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  2. Run the following command:

    gcloud ai endpoints explain ENDPOINT_ID \
      --region=LOCATION_ID \
      --json-request=request.json

    Replace the following:

    • ENDPOINT_ID: The ID for the endpoint.
    • LOCATION_ID: The region where you are using Vertex AI.

    Optionally, if you want to send an explanation request to a specific DeployedModel on the Endpoint, you can specify the --deployed-model-id flag:

    gcloud ai endpoints explain ENDPOINT_ID \
      --region=LOCATION \
      --deployed-model-id=DEPLOYED_MODEL_ID \
      --json-request=request.json

    In addition to the placeholders described previously, replace the following:

    • DEPLOYED_MODEL_ID Optional: The ID of the deployed model for which you want to get explanations. The ID is included in the predict method's response. If you need to request explanations for a particular model and you have more than one model deployed to the same endpoint, you can use this ID to ensure that the explanations are returned for that particular model.

REST

The following example shows an online prediction request for a tabular classification model with local feature attributions. The request format is the same for regression models.

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

  • LOCATION: Region where Endpoint is located. For example, us-central1.
  • PROJECT: Your project ID.
  • ENDPOINT_ID: The ID for the endpoint.
  • PREDICTION_DATA_ROW: A JSON object with keys as the feature names and values as the corresponding feature values. For example, for a dataset with a number, an array of strings, and a category, the row of data might look like the following example request:

    "length":3.6,
    "material":"cotton",
    "tag_array": ["abc","def"]
    

    A value must be provided for every feature included in training. The format of the data used for prediction must match the format used for training. Refer to Data format for predictions for details.

  • DEPLOYED_MODEL_ID (optional): The ID of the deployed model for which you want to get explanations. The ID is included in the predict method's response. If you need to request explanations for a particular model and you have more than one model deployed to the same endpoint, you can use this ID to ensure that the explanations are returned for that particular model.

HTTP method and URL:

POST https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/endpoints/ENDPOINT_ID:explain

Request JSON body:

{
  "instances": [
    {
      PREDICTION_DATA_ROW
    }
  ],
  "deployedModelId": "DEPLOYED_MODEL_ID"
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json, and execute the following command:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/endpoints/ENDPOINT_ID:explain"

PowerShell

Save the request body in a file named request.json, and execute the following command:

$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://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/endpoints/ENDPOINT_ID:explain" | Select-Object -Expand Content
 

Python

To learn how to install or update the Vertex AI SDK for Python, see Install the Vertex AI SDK for Python. For more information, see the Python API reference documentation.

def explain_sample(project: str, location: str, endpoint_id: str, instance_dict: Dict):

    aiplatform.init(project=project, location=location)

    endpoint = aiplatform.Endpoint(endpoint_id)

    response = endpoint.explain(instances=[instance_dict], parameters={})

    for explanation in response.explanations:
        print(" explanation")
        # Feature attributions.
        attributions = explanation.attributions
        for attribution in attributions:
            print("  attribution")
            print("   baseline_output_value:", attribution.baseline_output_value)
            print("   instance_output_value:", attribution.instance_output_value)
            print("   output_display_name:", attribution.output_display_name)
            print("   approximation_error:", attribution.approximation_error)
            print("   output_name:", attribution.output_name)
            output_index = attribution.output_index
            for output_index in output_index:
                print("   output_index:", output_index)

    for prediction in response.predictions:
        print(prediction)

Get explanations for a previously returned prediction

Because explanations increase resource usage, you might want to reserve requesting explanations for situations when you specifically need them. Sometimes, it can be helpful to request explanations for a prediction result you've already received, perhaps because the prediction was an outlier or did not make sense.

If all of your predictions are coming from the same model, you can simply resend the request data, with explanations requested this time. However, if you have multiple models returning predictions, you must make sure you send the explanation request to the correct model. You can view explanations for a particular model by including the deployed model's ID deployedModelID in your request, which is included in the response of the original prediction request. Note that the deployed model ID is different from the model ID.

Interpret explanation results

To calculate local feature importance, first the baseline prediction score is calculated. Baseline values are computed from the training data, using the median value for numeric features and the mode for categorical features. The prediction generated from the baseline values is the baseline prediction score. Baseline values are calculated once for a model and do not change.

For a specific prediction, the local feature importance for each feature tells you how much that feature added to or subtracted from the result as compared with the baseline prediction score. The sum of all of the feature importance values equals the difference between the baseline prediction score and the prediction result.

For classification models, the score is always between 0.0 and 1.0, inclusive. Therefore, local feature importance values for classification models are always between -1.0 and 1.0 (inclusive).

For examples of feature attribution queries and to learn more, see Feature Attributions for Classification and Regression.

Example output for predictions and explanations

Classification

The return payload for an online prediction from a tabular classification model with feature importance looks similar to the following example.

The instanceOutputValue of 0.928652400970459 is the confidence score of the highest-scoring class, in this case class_a. The baselineOutputValue field contains the baseline prediction score, 0.808652400970459. The feature that contributed most strongly to this result was feature_3.

{
"predictions": [
  {
    "scores": [
      0.928652400970459,
      0.071347599029541
    ],
    "classes": [
      "class_a",
      "class_b"
    ]
  }
]
"explanations": [
  {
    "attributions": [
      {
        "baselineOutputValue": 0.808652400970459,
        "instanceOutputValue": 0.928652400970459,
        "approximationError":  0.0058915703929231,
        "featureAttributions": {
          "feature_1": 0.012394922231235,
          "feature_2": 0.050212341234556,
          "feature_3": 0.057392736534209,
        },
        "outputIndex": [
          0
        ],
        "outputName": "scores"
      }
    ],
  }
]
"deployedModelId": "234567"
}

Regression

The return payload for an online prediction with feature importance from a tabular regression model looks similar to the following example.

The instanceOutputValue of 1795.1246466281819 is the predicted value, with the lower_bound and upper_bound fields providing the 95% confidence interval. The baselineOutputValue field contains the baseline prediction score, 1788.7423095703125. The feature that contributed most strongly to this result was feature_3.

{
"predictions": [
  {
    "value": 1795.1246466281819,
    "lower_bound": 246.32196807861328,
    "upper_bound": 8677.51904296875
  }
]
"explanations": [
  {
    "attributions": [
      {
        "baselineOutputValue": 1788.7423095703125,
        "instanceOutputValue": 1795.1246466281819,
        "approximationError": 0.0038215703911553,
        "featureAttributions": {
          "feature_1": 0.123949222312359,
          "feature_2": 0.802123412345569,
          "feature_3": 5.456264423211472,
        },
        "outputIndex": [
          -1
        ]
      }
    ]
  }
],
"deployedModelId": "345678"
}

What's next