Workflow Automation Templates
A library of ready-to-use workflow templates to accelerate your data journey
Data Drift
Measure data drift using ML metrics
Overview
This workflow detects and measures data drift using the H2O Distributed Random Forest (DRF) model and ML Data Metrics. It evaluates model performance across datasets to identify shifts in data distribution that may affect predictive accuracy.
Details
The process begins by loading the diabetes dataset and splitting it into training and testing subsets using the Split node. The H2O Distributed Random Forest node trains the model on the training data, which is then evaluated on the test data through the H2O Score node.
The ML Data Metrics node measures performance differences, highlighting potential data drift between datasets. Results are displayed using the Print N Rows node, and the trained model is saved via H2O ML Model Save for reuse.
This workflow helps monitor model stability, ensuring consistent predictive performance as data evolves over time.