Outlier detection is a great way to track anomalous input and output data. Verta monitoring system detects outliers for all input features and predictions. Along with outliers you can also monitor other quality metrics like mean, median, standard deviation over time.
The following data quality metrics are computed for all feature and prediction columns: outlier count, mean, median, min, max and standard deviation.
With the outlier chart (scatter plot) you can visualize outlier data points over time.
Note: The data points shown on charts are downsampled without impacting the correctness of the computations.
Copy link