I'm having a failure in the batch scoring pipeline. I created the storage account and can access it, but it's looking for input/diabetes_scoring_input.csv. I created the input container, but this csv is not in the repo. How does it get there? I'm unclear if this is a real problem or a documentation issue.