Datascientists and ML Engineers perform plenty of experimentations with their models. It becomes important to track all these experminents to conclude what kinda features and parameters suited the best performing model.
1. MLFlow helps in tracking the parameters and also stores the model artifact versions for each experiment. Makes easy to compare and conclude
2. Seldon Core helps in deploying the models at scale with A/B testing, Canary deployments
Google Slides for the above video
https://docs.google.com/presentation/d/1nh8uWlQHQ7yyKZPir8YZAFu3S1M1A7wu/edit?usp=sharing&ouid=113737295631185544437&rtpof=true&sd=true
