Lina Weichbrodt: What I learned from monitoring more than 30 Machine Learning Use Cases

Speaker:: Lina Weichbrodt Track: General: Production This talk summarizes the most important insights I gained from running more than 30 machine learning use cases in production. We will take a loan prediction model as an example use case and cover questions like: - What is the difference between metrics for model training and metrics for model monitoring? - Which metrics are generally useful to be monitored? - Which metrics should you prioritize? - How can monitoring be set up using a traditional software monitoring stack (tools like Grafana and Prometheus)? This talk will be useful for you if you: - are a hands-on engineer or data scientist - want to use your team’s or company’s existing monitoring and dashboarding infrastructure to monitor machine learning stacks - are a beginner or intermediate in MLOPs Recorded at the PyConDE & PyData Berlin 2022 conference, April 11-13 2022. More details at the conference page: Twitter: https:/