What is SageMaker Autopilot doing when in state "InProgress - AnalyzingData" ?

Question

Hi,

I'm trying this nice SageMaker Autopilot demo https://github.com/awslabs/amazon-sagemaker-examples/blob/master/autopilot/autopilot_customer_churn_high_level_with_evaluation.ipynb

At the beginning of the job, the status is "InProgress - AnalyzingData" for several minutes. This is long enough that I'd like to know more about it: what is Autopilot doing when at that status?

Accepted Answer

There are some metrics begin collected in this stage. To understanding what is doing is the same as what happens when you're using tensorflow autoML.  There's a deep explanation what is does in our Science page https://www.amazon.science/publications/amazon-sagemaker-autopilot-a-white-box-automl-solution-at-scale

What is SageMaker Autopilot doing when in state "InProgress - AnalyzingData" ?

Relevant content