Data science projects are not so straight-forward in practice. Many hands-on books describe various parts of the process in such a way that you may come to expect everything to go about smoothly and insights to naturally come out from your analyses. However, when you try to apply all this know-how in your own projects, you often end up with bugs in the code and making various mistakes in the process, without realizing it. Even though a colleague can help you pinpoint certain issues in your project, it is often the case that you need to find the majority of them on your own - and fix them. This video helps you get a better perspective of what are the most common problems you’ll encounter when working on a data science project. This know-how, along with some hands-on experience can help you eliminate these bugs that most data scientists face at one point or another, and make the whole process smoother and more enjoyable.
We will cover:
An overview of the data science process, including where bugs and mistakes are introduced and can be found
Types of programming bugs, including indexing and values issues
Common mistakes in the data science process, including sampling issues, identifying the wrong metrics, overfitting problems, and unnecessary assumptions
Strategies for coping with bugs, including documentation consultation and unit testing