AlertRule

AlertRule

Alert rule for automated monitoring and alerting in ML systems.

An AlertRule defines conditions that automatically trigger notifications when ML model metrics exceed specified thresholds. Alert rules are essential for proactive monitoring of model performance, data drift, and operational issues.

Example

# Create feature drift alert
drift_alert = AlertRule(
    name=”credit_score_drift”,
    model_id=model.id,
    metric_id=”drift_score”,
    priority=Priority.HIGH,
    compare_to=CompareTo.BASELINE,
    condition=AlertCondition.GT,
    bin_size=BinSize.HOUR,
    critical_threshold=0.8,
    warning_threshold=0.6,
    baseline_id=baseline.id,
    columns=[“credit_score”, “income”]
).create()

# Create performance degradation alert
perf_alert = AlertRule(
    name=”accuracy_drop”,
    model_id=model.id,
    metric_id=”accuracy”,
    priority=Priority.MEDIUM,
    compare_to=CompareTo.TIME_PERIOD,
    condition=AlertCondition.LT,
    bin_size=BinSize.DAY,
    critical_threshold=0.85,
    compare_bin_delta=7  # Compare to 7 days ago
).create()

# Configure notifications
drift_alert.set_notification_config(
    emails=[”[[e-mail]](mailto:[e-mail])”, “[[e-mail]](mailto:[e-mail])”],
    pagerduty_services=[“ML_ALERTS”],
    pagerduty_severity=”critical”
)

Alert rules continuously monitor metrics and trigger notifications when thresholds are exceeded. Use appropriate evaluation delays to avoid false positives from temporary data fluctuations.

Initialize an AlertRule instance.

Creates an alert rule configuration for automated monitoring of ML model metrics. The alert rule defines conditions that trigger notifications when thresholds are exceeded, enabling proactive monitoring of model performance and data quality.

Parameters

Parameter	Type	Required	Default	Description
name	str	✗	None	Human-readable name for the alert rule. Should be descriptive and unique within the model context.
model_id	UUID | str	✗	None	UUID of the model this alert rule monitors. Must be a valid model that exists in the Fiddler platform.
metric_id	str | UUID	✗	None	ID of the metric to monitor (e.g., “drift_score”, “accuracy”, “precision”, “recall”, custom metric IDs).
priority	Priority | str	✗	None	Alert priority level (HIGH, MEDIUM, LOW). Determines urgency and routing of notifications.
compare_to	CompareTo | str	✗	None	Comparison method for threshold evaluation: BASELINE: Compare against a fixed baseline; TIME_PERIOD: Compare against previous time period; RAW_VALUE: Compare against absolute threshold
condition	AlertCondition | str	✗	None	Alert condition (GT, LT, OUTSIDE_RANGE). Defines when the alert should trigger relative to the threshold.
bin_size	BinSize | str	✗	None	Time aggregation window (HOUR, DAY, WEEK). Controls how data is grouped for metric calculation.
threshold_type	AlertThresholdAlgo | str	✗	None	Threshold calculation method (MANUAL or AUTO). MANUAL uses user-defined thresholds, AUTO calculates dynamic thresholds based on historical data.
auto_threshold_params	dict[str, Any] | None	✗	None	Parameters for automatic threshold calculation. Used when threshold_type is AUTO.
critical_threshold	float | None	✗	None	Critical alert threshold value. Triggers high-priority notifications when exceeded.
warning_threshold	float | None	✗	None	Warning alert threshold value. Triggers medium-priority notifications when exceeded.
columns	list[str] | None	✗	None	List of feature columns to monitor. For feature-specific drift alerts. If None, monitors all features.
baseline_id	UUID | str | None	✗	None	UUID of the baseline to compare against. Required when compare_to is BASELINE.
segment_id	UUID | str | None	✗	None	UUID of the data segment to monitor. For segment-specific monitoring (optional).
compare_bin_delta	int | None	✗	None	Number of time bins to compare against. Used with TIME_PERIOD comparison (e.g., 7 for week-over-week).
evaluation_delay	int	✗	None	Delay in minutes before evaluating alerts. Helps avoid false positives from incomplete data.
category	str | None	✗	None	Custom category for organizing alerts. Useful for grouping related alerts in dashboards.

Example

# Feature drift alert with baseline comparison
drift_alert = AlertRule(
    name=”income_drift_detection”,
    model_id=model.id,
    metric_id=”drift_score”,
    priority=Priority.HIGH,
    compare_to=CompareTo.BASELINE,
    condition=AlertCondition.GT,
    bin_size=BinSize.HOUR,
    critical_threshold=0.8,
    warning_threshold=0.6,
    baseline_id=baseline.id,
    columns=[“income”, “credit_score”],
    evaluation_delay=15,  # 15 minute delay
    category=”data_quality”
)

# Performance monitoring with time comparison
perf_alert = AlertRule(
    name=”weekly_accuracy_check”,
    model_id=model.id,
    metric_id=”accuracy”,
    priority=Priority.MEDIUM,
    compare_to=CompareTo.TIME_PERIOD,
    condition=AlertCondition.LT,
    bin_size=BinSize.DAY,
    critical_threshold=0.85,
    compare_bin_delta=7,  # Compare to 7 days ago
    category=”performance”
)

After initialization, call create() to persist the alert rule to the Fiddler platform. Alert rules begin monitoring immediately after creation.

classmethod get(id_)

Retrieve an alert rule by its unique identifier.

Fetches an alert rule from the Fiddler platform using its UUID. This method returns the complete alert rule configuration including thresholds, notification settings, and monitoring status.

Parameters

• id – The unique identifier (UUID) of the alert rule to retrieve. Can be provided as a UUID object or string representation.

• id_ (UUID | str)

Returns

The alert rule instance with all configuration and metadata populated from the server.

Return type: AlertRule

Raises

• NotFound – If no alert rule exists with the specified ID.

• ApiError – If there’s an error communicating with the Fiddler API.

Example

# Retrieve alert rule by ID
alert_rule = AlertRule.get(id_=”550e8400-e29b-41d4-a716-446655440000”)
print(f”Alert: {alert_rule.name}”)
print(f”Metric: {alert_rule.metric_id}”)
print(f”Priority: {alert_rule.priority}”)
print(f”Critical threshold: {alert_rule.critical_threshold}”)

# Check notification configuration
notification_config = alert_rule.get_notification_config()
print(f”Email recipients: {notification_config.emails}”)

This method makes an API call to fetch the latest alert rule configuration from the server, including any recent threshold or notification updates.

classmethod list(model_id, metric_id=None, columns=None, baseline_id=None, ordering=None)

Get a list of all alert rules in the organization.

Parameters

Parameter	Type	Required	Default	Description
model_id	UUID | str	✗	None	list from the specified model
metric_id	UUID | str | None	✗	None	list rules set on the specified metric id
columns	list[str] | None	✗	None	list rules set on the specified list of columns
baseline_id	UUID | str | None	✗	None	list rules set on the specified baseline_id
ordering	list[str] | None	✗	None	order result as per list of fields. [“-field_name”] for descending

Returns

paginated list of alert rules for the specified filters

Return type: Iterator[AlertRule]

delete()

Delete an alert rule.

Return type: None

create()

Create a new alert rule.

Return type: AlertRule

update()

Update an existing alert rule.

Return type: None

enable_notifications()

Enable notifications for an alert rule

Return type: None

disable_notifications()

Disable notifications for an alert rule

Return type: None

set_notification_config()

Set notification config for an alert rule

Parameters

Parameter	Type	Required	Default	Description
emails	list[str] | None	✗	None	list of emails
pagerduty_services	list[str] | None	✗	None	list of pagerduty services
pagerduty_severity	str | None	✗	None	severity of pagerduty
webhooks	list[UUID] | None	✗	None	list of webhooks UUIDs

Returns

NotificationConfig object

Return type: NotificationConfig

get_notification_config()

Get notifications config for an alert rule

Returns

NotificationConfig object

Return type: NotificationConfig