{"id":305,"date":"2021-04-08T14:41:49","date_gmt":"2021-04-08T22:41:49","guid":{"rendered":"https:\/\/live-cometml.pantheonsite.io\/?p=305"},"modified":"2025-04-24T17:30:31","modified_gmt":"2025-04-24T17:30:31","slug":"introducing-comet-mpm-model-production-monitoring","status":"publish","type":"post","link":"https:\/\/www.comet.com\/site\/blog\/introducing-comet-mpm-model-production-monitoring\/","title":{"rendered":"Introducing Comet MPM: Model Production Monitoring"},"content":{"rendered":"\n

For many of us, it\u2019s already a struggle to take a seemingly successful ML model live, but deployment is only half the battle. There are factors we can\u2019t foresee that make our \u201cbest\u201d models fall short of expectations. Maybe we haven\u2019t seen what production data looks like or our model needs to be updated more frequently than we expected. Regardless of reason, it\u2019s not enough to track hyperparameters and metrics, we also have to monitor how our models work after they\u2019ve been deployed to production.<\/p>\n\n\n\n

Without tools to understand how your model is running in a production environment, it can be difficult to compare to expected model results \u2013 much less be able to iterate on those models after they\u2019ve hit production.<\/p>\n\n\n\n

That\u2019s where production model monitoring comes in. It\u2019s the starting point for understanding how your production models are performing in real time, which enables you to do things like check for accuracy drift, the phenomena of model accuracy slowly drifting away from expected results.<\/p>\n\n\n\n

That\u2019s why we\u2019re excited to announce Comet MPM (Model Production Monitoring)<\/a> \u2013 a new product that will enable you to get visibility into your model performance in real time.<\/p>\n\n\n\n

What is Comet MPM?<\/h2>\n\n\n\n

As data scientists, our work doesn\u2019t stop when you\u2019ve built and chosen some candidate models. Active production models deserve the same care and attention, because machine learning projects can fail at any step of each model\u2019s life cycle.<\/p>\n\n\n\n

This is where Comet MPM comes in \u2013 it helps to answer questions like:<\/p>\n\n\n\n