MLOps with MLflow скачать в хорошем качестве

MLOps with MLflow

Today's Data Scientists (and budding ones too) build ML models with tons of tools and libraries at their disposal. Not many have exposure to MLOps. Wikipedia defines MLOps as "a set of practices that aims to deploy and maintain machine learning models in production reliably and efficiently." This tech talk is aimed at imparting that knowledge with MLfLow as a tool to manage ML pipelines and deploy ML models in production. MLflow is an open source platform to manage the ML lifecycle, including experimentation, reproducibility, deployment, and a central model registry. This will be a hands-on session that focuses on building, managing and deploying models using MLflow APIs and serve models as microservices. Any model building experience (even textbook exercises) is sufficient to follow the session. Speaker: Ramanathan R M, Sr. Architect (Data Science) @ SLK Software Code: https://github.com/RaamRaam/mlflow_we...