I'm using Azure Custom Vision to create an app that will inform me whether or not a certain product is defective. For that, I used the Custom Vision website to train a model. The training is complete, and the model is ready to use. I get projected labels with percentages of the components in the photo when I submit a picture of the object. My issue now is: how can I obtain the data for this proportion in order to work with it? I need them to be implemented in Visual Studio so that I can develop C# code that allows the AI to determine whether an object is defective or not.