Expecting constant change
Everything is in a constant state of change. 15 years ago, only a few people ever heard about neural networks and machine learning. Today, you have access to a vast amount of ML libraries, programs, and cloud services. Every day, new progress is made to automate ML tasks and improve ML modeling. Just think about the voice assistants you may use and what is happening with self-driving vehicles.
Due to this, you are in for a whole bunch of constant changes being made to ML libraries and their tooling. This is especially true in a cloud environment, where updates can quickly be pushed out to the userbase compared to licensed software. As we learned previously, looking at the big cloud providers, their services can typically be divided into the following categories:
- Infrastructure as a Service (IaaS): IaaS services are all-infrastructure abstractions such as virtual machines (compute), disks (storage), and networking.
- Platform as a Service (PaaS...