aws-samples / amazon-sagemaker-pipeline-deploy-manage-100x-models-python-cdk Goto Github PK

when new data added and new endpoint created , is new end point created with the same name as old point
since it end point used in lambda function to access end point from local computer
then end point name should not be changed

but by default , for sagemaker new end point can not be created with the same name as old point
and end point configuration should be deleted

naming convention for new end point should be the same for updated endpoint for using in production with Lambda function

as you wrote
Start SM Pipeline Stack: Designed to respond to new training data uploaded to the specified S3 bucket. It utilizes a Lambda function to trigger the SageMaker pipeline, ensuring that your machine learning models are updated with the latest data automatically

This approach allows for the creation and training of 100x different models, with each model associated with a distinct DummyDimension.

may you clarify what is it 100x different models
do you have video for this project how to use?

folder DummyDim4 not existing but references

key="training-dataset/DummyDim4/churn_clubbed.csv"

may help to understand where in your code you use files for training
https://github.com/aws-samples/amazon-sagemaker-pipeline-deploy-manage-100x-models-python-cdk/tree/main/sagemaker_pipeline_deploy_manage_n_models_cdk/dataset/training-dataset