We all use default values in our code: it could be 0, it could be -1. Maybe your default value of choice is NaN or something else. Code, and especially the model, assume that the default value will not change.Â
When a default value is changed without the relevant parties knowing, for example, the data science team, it can result in unexpected inputs and unexpected behavior of a model.
Learn more about how this can affect your model in production.Â