When I reviewed Amazon SageMaker in 2018, I noted that it was a extremely scalable device studying and deep studying service that supports 11 algorithms of its individual, in addition any other individuals you supply. Hyperparameter optimization was nonetheless in preview, and you essential to do your individual ETL and aspect engineering.
Considering the fact that then, the scope of SageMaker has expanded, augmenting the core notebooks with IDEs (SageMaker Studio) and automated device studying (SageMaker Autopilot) and including a bunch of essential companies to the overall ecosystem, as proven in the diagram beneath. This ecosystem supports device studying from preparing as a result of model building, instruction, and tuning to deployment and management — in other words, close to close.
What’s new in SageMaker?
What’s new? Specified that I last seemed at SageMaker just after it was unveiled, the record is relatively lengthy, but let us commence with the most noticeable companies.