sagemaker-mlops-end-to-end | And end-to-end demo of MLOps in SageMaker

IF NOT USING FEATURE STORE, IGNORE THE STEPS ABOVE AND FOLLOW THE BELOW STEPS. To setup Model Monitor. Once all the Endpoints have been created, navigate to the Endpoint UI in SageMaker Studio. Click on the endpoint deployed using the notebook in the model monitor folder,. Things to highlight in the demo -.
If using Feature Store, the first step of the Pipeline will need to read data from Feature Store.
In the file sagemaker-pipeline/pipeline-dw-fs.py lines 131 to 178 need to be replaced with the code in the notebook created by the DataWrangler Export.
The first step of the Pipeline will be step_read_train
Replace the first step in sagemaker-pipeline/pipeline.py with step_read_train and step_process from sagemaker-pipeline/pipeline-dw-fs.py.
Navigate to the model build repo created by the SageMaker Project, replace the code in pipelines/abalone/ with the code in sagemaker-pipeline/.
Trigger the pipeline by pushing the new code to the CodeCommit/Git repo (depending on the template selected)
Once the pipeline has completed, find the model package group in the Model Registry and find the ARN of the model package created in the group
Approve the model in the model registry, this will trigger the model deployment pipeline, you should see an endpoint being created in SageMaker
This endpoint will have the suffix -staging. You can navigate to CodePipeline, and under Pipelines you will see one with your project name and model-deploy. Click on that Pipeline and you will see a manual approval option. When approved, a new endpoint will be created with the suffix -prod.
These endpoints are created by the default seed code in the 1st party template and do not have Data Capture enabled.
Navigate to model-monitor/create_endpoint.ipynb to create an endpoint with DataCapture enabled
Run model-monitor/data_quality_monitor.ipynb to set up a Data Quality Monitoring schedule on the endpoint.
End to end lineage View of the trial component from the model in the Model Registry Lineage from the Endpoint to the Model Package Group and Version
Pipelines integration with Experiments
Debugging a Pipeline through the DAG view
CI/CD for automatic training and deployment