Invoke SageMaker Endpoint action
- Updated: 2023/11/02
Invoke SageMaker Endpoint action
Invoke SageMaker Endpoint action sends data to an AI model deployed on the SageMaker platform and receives response. The action enables you to integrate machine learning capabilities directly into your Control Room, making predictions or classifications based on the trained model's expertise.
Note: The examples and graphics provided on this page are
for representation purposes only and may not accurately reflect your specific instance. We do
not assume responsibility for their maintenance or accuracy.
Overview
- Prepare Data: Before invoking the endpoint, you need to prepare the data you want to send for prediction. This could be an image, a text snippet, numerical values, or any other format hat the model expects.
- Invoke Endpoint: To get predictions, you send a POST request to the endpoint URL with the prepared data as input. The request typically includes the data in JSON format.
- Endpoint Processing: The SageMaker endpoint processes the incoming data using the deployed model. It runs the data through the model's algorithms and produces predictions or responses based on the model's training.
- Response: The endpoint sends back the model's predictions or responses in the HTTP response body. Your application can then parse this response and use the results as needed.
Prerequisites
- The AI model should be deployed in SageMaker
- The deployed model is accessible by Invoke SageMaker Endpoint.
Note: This task is usually performed by an automation admin with
Admin privileges.
Procedure
-
Click Management Console corresponding to the project
you selected.
-
Search for SageMaker and select Amazon
SageMaker.
-
Navigate to the Domains and click a Domain.
Alternately, you can click Studio (on the left navigation) to select the
appropriate domain and click Open Studio. To add a
domain, see AWS add a custom domain. For more information on
Domains, see Amazon SageMaker Domain.