Model Versioning and Webhooks

In the previous chapter, we delved deep into the capabilities of Databricks AutoML, exploring its various components in detail. We gained a comprehensive understanding of how data science practitioners can harness the power of transparent “glass box” AutoML to kickstart their machine learning solutions seamlessly, especially when tackling complex business challenges.

Furthermore, we put AutoML into action by automating the selection of a candidate model for our Bank Customer Churn prediction classification problem. To facilitate this process, we seamlessly integrated the robust MLflow features into our workflow. This integration allowed us to meticulously track every aspect of our model’s training, providing us with invaluable insights into its performance and enabling us to make data-driven decisions. Our journey also took us to the MLflow tracking server, where we logged and monitored the entire training process, ensuring that...