Incorporate SageMaker Data Wrangler into SageMaker Pipelines

Yes, it is possible to incorporate SageMaker Data Wrangler as a step in SageMaker Pipelines.

• SageMaker Pipelines allows you to define and execute a sequence of ML workflow steps such as data preprocessing, model training, evaluation etc.
• SageMaker Data Wrangler can be used to prepare and transform data. It provides notebooks, flows and processing jobs capabilities.
• To add a Data Wrangler step in a Pipeline, you can define a SageMaker Processing job that runs a Data Wrangler notebook or flow.
• This processing job can be configured as the first step in the Pipeline. It will run the data wrangling tasks to preprocess the data.
• The output of the Data Wrangler processing job can then be used as input to the subsequent training step in the Pipeline.
• This allows automated execution of end-to-end ML workflows with data preparation via Data Wrangler followed by model training/evaluation via Pipelines.

Yes. It is possible. Please refer the following links for additional details.

https://aws.amazon.com/blogs/machine-learning/integrate-amazon-sagemaker-data-wrangler-with-mlops-workflows/

https://github.com/aws-samples/sm-data-wrangler-mlops-workflows