Matlab model version management

Hi, We have now model version 51. The model processes various risk factors, sometime we don't want to process some risk factors so we fall back to a model version which excludes such risk factors. For example Version 51 processes 10 risk factors but version 36 processes 8. If we want to see distributions of only these 8 risk factors we run code version 36. I don't think this is a good practice of engineering the model and managing the code. Some argue that this eases testing techniques and the proponent are in favor of such code version management. I am an opponent of such practice, I am not a matlab developer myself and don't know how to address this and prove my point. Do you have any best practices and guidelines for such a situation. Thanks for help Ranjan

Answer 1

[Yuebin Zhou]

Version management system should only be used for managing versions. As you have pointed out, using it in this case may not be a good practice.