While creating sagemaker baseline how to form the json input ?

Question

Here(https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-create-baseline.html) it is mentioned "Note that the prediction/output columns are assumed to be the first columns in the training dataset."
If our input is flat JSON how to specify the "prediction/output" ?

Accepted Answer

looks like it will be follow the pattern below

Input - Flattened JSON {"feature0": , "feature1": ...} - Tabular ", ..." - Output: - Flattened JSON {"prediction0": , "prediction1": ...} - Tabular ", ..."

Refer -> (https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-pre-and-post-processing.html#model-monitor-pre-processing-script)  " For example, suppose the output of your model is an array [1.0, 2.1]. The Amazon SageMaker Model Monitor container only works with tabular or flattened JSON structures, like {“prediction0”: 1.0, “prediction1” : 2.1}. You could use a preprocessing script like the following to transform the array into the correct JSON structure."

While creating sagemaker baseline how to form the json input ?

Conteúdo relevante