Hands-On Design Patterns and Best Practices with Julia

Adapting to growth

No matter what kind of application you develop, it should not be hard to recognize growing pains. 

For a data science project, the following signs typically indicate a growth-related problem:

• "My notebook is getting too long. I often have to scroll up and down to understand what I have done before and what I'm doing now. There are too many variables created in between and I'm losing track of what they mean and how they are used."
• "The data structure is too complex. I was working on a data frame and have transformed it in ten different ways. I have now lost track of which transformed version represents what, and why they were needed in the first place."
• "I have saved a bunch of machine learning models on disk, and I'm losing ...