You create a binary classification model. The model is registered in an Azure Machine Learning workspace. You use the Azure Machine Learning Fairness SDK to assess the model fairness. You develop a…

Question

Options

• AImplement the download_dashboard_by_upload_id function
• BImplement the function
• CImplement the upload_dashboard_dictionary function
• DUpload the training script

• A
  9% (2)
• B
  5% (1)
• C
  82% (18)
• D
  5% (1)

Explanation

You should implement the upload_dashboard_dictionary function from the azureml.contrib.fairness package. When model fairness metrics are computed locally using the Fairlearn SDK, the results are stored as a dashboard dictionary object. The upload_dashboard_dictionary() function serializes this dictionary and uploads it to the Azure ML workspace, making the fairness metrics visible in Azure ML Studio's Fairness dashboard. The download_dashboard_by_upload_id function (A) retrieves an already-uploaded dashboard rather than uploading new data. Uploading just the training script (D) does not transfer the computed fairness metrics to Studio.

No community discussion yet for this question.