Schema for Violations (constraint_violations.json file) - Amazon SageMaker AI

Schema for Violations (constraint_violations.json file)

The violations file is generated as the output of a MonitoringExecution, which lists the results of evaluating the constraints (specified in the constraints.json file) against the current dataset that was analyzed. The Amazon SageMaker Model Monitor prebuilt container provides the following violation checks.

{
    "violations": [{
      "feature_name" : "string",
      "constraint_check_type" :
              "data_type_check",
            | "completeness_check",
            | "baseline_drift_check",
            | "missing_column_check",
            | "extra_column_check",
            | "categorical_values_check"
      "description" : "string"
    }]
}

Types of Violations Monitored

Violation Check Type Description
data_type_check

If the data types in the current execution are not the same as in the baseline dataset, this violation is flagged.

During the baseline step, the generated constraints suggest the inferred data type for each column. The monitoring_config.datatype_check_threshold parameter can be tuned to adjust the threshold on when it is flagged as a violation.
completeness_check

If the completeness (% of non-null items) observed in the current execution exceeds the threshold specified in completeness threshold specified per feature, this violation is flagged.

During the baseline step, the generated constraints suggest a completeness value.

baseline_drift_check

If the calculated distribution distance between the current and the baseline datasets is more than the threshold specified in monitoring_config.comparison_threshold, this violation is flagged.
missing_column_check

If the number of columns in the current dataset is less than the number in the baseline dataset, this violation is flagged.
extra_column_check

If the number of columns in the current dataset is more than the number in the baseline, this violation is flagged.
categorical_values_check

If there are more unknown values in the current dataset than in the baseline dataset, this violation is flagged. This value is dictated by the threshold in monitoring_config.domain_content_threshold.