Chapter 1. Introduction to MLOps
Since 1986, I have had a few more deaths, several from insufficient attention but mainly from deliberately pushing the limits in various directions—taking a chance in bonsai is a bit like taking a chance with love; the best outcome requires risky exposure to being hurt and no guarantee of success.
Dr. Joseph Bogen
One of the powerful aspects of science fiction is its ability to imagine a future without constraints. One of the most influential science fiction shows of all time is the TV show Star Trek, which first aired in the mid-1960s, approximately 60 years ago. The cultural impact inspired designers of technology like the Palm Pilot and hand-held cellular phones. In addition, Star Trek influenced the cofounder of Apple computer, Steve Wozniak, to create Apple computers.
In this age of innovation in machine learning, there are many essential ideas from the original series relevant to the coming MLOps (or Machine Learning Operations) industrial revolution. For example, Star Trek hand-held tricorders can instantly classify objects using pretrained multiclass classification models. But, ultimately, in this futuristic science-fiction world, domain experts like the science officers, medical officers, or captain of the ship, don’t spend months training machine learning models. Likewise, the crew of their science vessel, the Enterprise, are not called data scientists. Instead, they have jobs where they often use data science.
Many of ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access