Deployng a Keras model with KServe (formerly KFServing) and EKS
- Here's a detailed step-by-step guide for doing it: guide.md
- Want to talk about it? Join DataTalks.Club
- To learn more about machine learning and deploying models check Machine Leanring Zoomcamp - a free course about machine leanring engineering
This tutorial was updated in December 2021 to reflect latest changes:
- KFServing is no longer a part of Kubeflow ecosystem and now is called KServe
- The version if KServe used here is 0.7.0
- The api version of yaml config files is
serving.kserve.io/v1beta1