Fairness
Fairness is a topic that we need to investigate in use cases where people are involved. Proper precautions should be taken to figure out whether the dataset is fair or not. In Azure ML, with the Responsible AI Toolbox, we can create a fairness dashboard. To do that, first, we need to know which features need to be fair, such as sex and race. Once we know that, we can create a dashboard, as shown in Figure 8.24.
In this section, we are going to create a fairness dashboard with a sample dataset:
- Go to the Azure ML Studio UI.
- Start the compute instance.
- Click on Notebook in the Author section.
- Create a new notebook with Python 3.8 with Azure ML as the kernel.
- Create a new notebook called
FairnessDashboard. - Import all the required libraries:
Figure 8.24 – Fairness imports
- Next, we are going to load the sample dataset:
Figure 8.25 – Getting the sample data
-
...
