Successfully building and deploying a machine-learning model can be difficult to do once. Enabling other data scientists (or yourself) to reproduce your pipeline, compare the results of different versions, track what’s running where, and redeploy and rollback updated models is much harder. In this eBook, we’ll explore what makes the ML lifecycle so challenging compared to the traditional software development lifecycle, and share how to address these challenges with Azure Databricks.
Standardizing the Machine Learning Lifecycle
Este recurso está disponible en English.