New Course on Machine Learning Operations
Building ML from scratch and deploying it in production
Preview of “Introduction to MLOps: Taking Machine Learning from Scratch to Production
Most people start learning Machine Learning (ML) with courses that teach the basics of building ML models and solutions. My first course “Learn AI With A Robot” was aimed at teaching basic ML concepts with the help of a robot.
In my professional life, I have found it much easier to learn backwards. Learning backwards first implies thinking about a crisp problem that needs to be solved, and then acquiring all the skills that are needed to solve the problem. Specifically for Machine Learning, the crisp problem is to successfully deploy an agent which can perform a concrete functionality with the help of Machine Learning techniques. There are relatively few educational materials which teach you how to create and bootstrap a Machine Learning based solution from scratch. Hence, I am embarking on an effort to develop a course which will teach this.
Taking Machine Learning (ML) Scratch to Production
In Taking ML From Scratch to Production, we consider how to deploy a Machine Learning solution to allow a Vector robot to identify another Vector robot using its camera. We discuss how to bootstrap an ML solution and then manage it in production with two industry standard services: Roboflow and ModelDB. We will discuss the operation workflow of Uploading Images → Annotating them →Image transformations and augmentation → Training a model → Maintaining a number of models and updating the model to use in production → Running Inference in production.
Available Exclusively at LearnWithARobot.com
I am still in the process of developing this course. All episodes of the course will be posted exclusively at LearnWithARobot.com If you are not a subscriber, please consider signing up. Meanwhile, please enjoy the introductory video at the top of this post.