Celebrate Py Day!
By Kjell Carlsson2022-03-143 min read
Pi Day is upon us and, while we celebrate “Pi” with pie, we mustn’t forget that other great “Pi” in our lives, which is, of course, Python! That programming language which has emerged as the de facto standard for production-grade data science, used by the majority of production-grade data scientists.
Take a moment to reflect on Python. Why do we love it? Why has it become so popular, so quickly, for data science? There are lots of reasons, but three stand out in particular.
3 Reasons to Love Python
- Python is collaboration. Yes, that includes collaboration between data scientists, data engineers and developers who use it as a common language, and the fact that Python is open source. More important by far is the collaboration it facilitates across different tools. Python is a superb wrapper not just for its own data science-focused libraries, but also for every new ML or AI tool or framework. Odds are, if you can’t leverage something using Python, either it’s under development, or it's just not meant for production data science.
- Python is iteration. Data science is all about iteration. You’ve got to iterate through every part of the model lifecycle: feature engineering, algorithm selection, hyperparameter tuning, validation, and even model monitoring. Python beats visual tools for data science because it is inherently easier to iterate using code. Just ask any developer. However, Python also wins over other programming languages because it is an interpreted programming language. Consequently, it can be stopped to experiment (or debug) anywhere in the development process.
- Python is production. Python isn’t just a wrapper for different data science tools, it is a wrapper for your systems across the model life cycle. It acts like a swiss army knife that enables you to stitch together your data from your data lake or data warehouse, your methods from your framework(s) of choice, and the systems that you will use to deploy your model, whether those be on premises or in the cloud. Python allows you to glue these together, while at the same time filling in any missing gaps.
That’s not to say Python is perfect. You need to support it with security, governance, productivity tools, and scalable infrastructure using an enterprise-grade platform (such as the Domino Enterprise MLOps Platform, which represents the gold standard for enterprise data science using Python). But we wouldn’t be in an “AI revolution” if it weren’t for Python. So, my fellow data scientists, take this opportunity to make Pi day your own and reflect on how Python has made your impact on the world —not to mention your value to your organization— what it is today.
Kjell Carlsson is the head of AI strategy at Domino Data Lab where he advises organizations on scaling impact with AI technologies. Previously, he covered AI, ML, and data science as a Principal Analyst at Forrester Research. He has written dozens of reports on AI topics ranging from computer vision, MLOps, AutoML, and conversation intelligence to augmented intelligence, next-generation AI technologies, and data science best practices. He has spoken in countless keynotes, panels, and webinars, and is frequently quoted in the media. Dr. Carlsson is also the host of the Data Science Leaders podcast and received his Ph.D. from Harvard University.
Subscribe to the Domino Newsletter
Receive data science tips and tutorials from leading Data Science leaders, right to your inbox.