Arize AI
Search…
User Guides
On-Premise Deployment
Data Quality Monitors
Overview
Maintaining the quality of data flowing into a model can be challenging due to upstream data changes and the increasing pace of data proliferation as organizations scale. Data quality monitors enable teams to quickly catch when features, predictions, or actuals data don't conform with what is expected.
Uses Include:
- Verify feature data is being sent in / not missing
- Catch when data deviates from a specified range or surpasses an accepted threshold
- Detect extreme model inputs or outputs
Setting Up Data Quality Monitors:
All monitors, including data quality, can be set up directly from the Monitors Page or from any Model Overview page. To learn more visit the Set up Model Monitors page.
.gif?alt=media&token=b2b35ae8-4aa3-4830-84fa-0fac4a6e6231)
Click 'Create' to begin configuring a new data quality monitor. Monitors can be set to the exact parameters you want to monitor for any model, model version, and dimension. In the example below, the monitor for the model wine_quality is set to trigger if the feature alcohol dips below 11.5%:
Data Quality Monitor Configuration
Once you are finished configuration the monitor, click 'Save Changes' and you're all set! You will now be able to track the health of the new monitor in the platform; and if you set up email notifications, receive alerts when it is triggered.
Last modified 3mo ago
Copy link