2 個答案
- 最新
- 最多得票
- 最多評論
0
One way is to use S3 trigger when AutoML job saves artifacts to a specified S3 bucket at job completion: https://docs.aws.amazon.com/AmazonS3/latest/userguide/EventBridge.html
There is also a processing job state change in SageMaker that you can connect with EventBridge as well: https://docs.aws.amazon.com/sagemaker/latest/dg/automating-sagemaker-with-eventbridge.html#processing-job-state
已回答 1 個月前
0
The solution for me was to trigger the lambda for all event sourced from sagemaker, and describe the auto ml job to check if it is completed. I had to add a sleep of 10 seconds when I receive the last event otherwise the auto ml job is not yet on a completed state. Hope AWS can create this event in the future.
已回答 1 個月前
相關內容
- 已提問 1 年前
- AWS 官方已更新 7 個月前
- AWS 官方已更新 1 年前
- AWS 官方已更新 2 年前
- AWS 官方已更新 1 年前
I am exploring a solution by processing all the event bridge events from source aws.sagemaker, this lambda requests the API to describe the auto ML job to understand if "AutoMLJobStatus": "Completed" , but the last event I receive I still do not have the completed status. I tried with only processing job state and I still can't get the final event when the auto ml is completed. I wish I had a specific event for autopilot since we have that status on describe request api.