feat: Add notification_channels field to model monitoring alert config.

PiperOrigin-RevId: 593812234

Author: vertex-sdk-bot

google/cloud/aiplatform/model_monitoring/__init__.py

@@ -15,17 +15,23 @@
 # limitations under the License.
 #
 
-from google.cloud.aiplatform.model_monitoring.alert import EmailAlertConfig
+from google.cloud.aiplatform.model_monitoring.alert import (
+    AlertConfig,
+    EmailAlertConfig,
+)
 from google.cloud.aiplatform.model_monitoring.objective import (
     SkewDetectionConfig,
     DriftDetectionConfig,
     ExplanationConfig,
     ObjectiveConfig,
 )
-from google.cloud.aiplatform.model_monitoring.sampling import RandomSampleConfig
+from google.cloud.aiplatform.model_monitoring.sampling import (
+    RandomSampleConfig,
+)
 from google.cloud.aiplatform.model_monitoring.schedule import ScheduleConfig
 
 __all__ = (
+    "AlertConfig",
     "EmailAlertConfig",
     "SkewDetectionConfig",
     "DriftDetectionConfig",

google/cloud/aiplatform/model_monitoring/alert.py

@@ -15,56 +15,80 @@
 # limitations under the License.
 #
 
-from typing import Optional, List
+from typing import List, Optional
 from google.cloud.aiplatform_v1.types import (
     model_monitoring as gca_model_monitoring_v1,
 )
 
-# TODO(b/242108750): remove temporary logic once model monitoring for batch prediction is GA
+# TODO(b/242108750): remove temporary logic once model monitoring for
+# batch prediction is GA.
 from google.cloud.aiplatform_v1beta1.types import (
     model_monitoring as gca_model_monitoring_v1beta1,
 )
 
 gca_model_monitoring = gca_model_monitoring_v1
 
 
-class EmailAlertConfig:
+class AlertConfig:
     def __init__(
-        self, user_emails: List[str] = [], enable_logging: Optional[bool] = False
+        self,
+        user_emails: List[str] = [],
+        enable_logging: Optional[bool] = False,
+        notification_channels: List[str] = [],
     ):
-        """Initializer for EmailAlertConfig.
+        """Initializer for AlertConfig.
 
         Args:
-            user_emails (List[str]):
-                The email addresses to send the alert to.
-            enable_logging (bool):
-                Optional. Defaults to False. Streams detected anomalies to Cloud Logging. The anomalies will be
-                put into json payload encoded from proto
-                [google.cloud.aiplatform.logging.ModelMonitoringAnomaliesLogEntry][].
-                This can be further sync'd to Pub/Sub or any other services
-                supported by Cloud Logging.
+            user_emails (List[str]): The email addresses to send the alert to.
+            enable_logging (bool): Optional. Defaults to False. Streams detected
+              anomalies to Cloud Logging. The anomalies will be put into json
+              payload encoded from proto
+              [google.cloud.aiplatform.logging.ModelMonitoringAnomaliesLogEntry][].
+              This can be further sync'd to Pub/Sub or any other services supported
+              by Cloud Logging.
+            notification_channels (List[str]): The Cloud notification channels to
+              send the alert to.
         """
-        self.enable_logging = enable_logging
         self.user_emails = user_emails
+        self.enable_logging = enable_logging
+        self.notification_channels = notification_channels
         self._config_for_bp = False
 
-    # TODO(b/242108750): remove temporary logic once model monitoring for batch prediction is GA
     def as_proto(self) -> gca_model_monitoring.ModelMonitoringAlertConfig:
-        """Converts EmailAlertConfig to a proto message.
+        """Converts AlertConfig to a proto message.
 
         Returns:
-            The GAPIC representation of the email alert config.
+            The GAPIC representation of the alert config.
         """
+        # TODO(b/242108750): remove temporary logic once model monitoring for
+        # batch prediction is GA.
         if self._config_for_bp:
             gca_model_monitoring = gca_model_monitoring_v1beta1
         else:
             gca_model_monitoring = gca_model_monitoring_v1
-        user_email_alert_config = (
-            gca_model_monitoring.ModelMonitoringAlertConfig.EmailAlertConfig(
-                user_emails=self.user_emails
-            )
-        )
+
         return gca_model_monitoring.ModelMonitoringAlertConfig(
-            email_alert_config=user_email_alert_config,
+            email_alert_config=gca_model_monitoring.ModelMonitoringAlertConfig.EmailAlertConfig(
+                user_emails=self.user_emails
+            ),
             enable_logging=self.enable_logging,
+            notification_channels=self.notification_channels,
         )
+
+
+class EmailAlertConfig(AlertConfig):
+    def __init__(
+        self, user_emails: List[str] = [], enable_logging: Optional[bool] = False
+    ):
+        """Initializer for EmailAlertConfig.
+
+        Args:
+            user_emails (List[str]): The email addresses to send the alert to.
+            enable_logging (bool): Optional. Defaults to False. Streams detected
+              anomalies to Cloud Logging. The anomalies will be put into json
+              payload encoded from proto
+              [google.cloud.aiplatform.logging.ModelMonitoringAnomaliesLogEntry][].
+              This can be further sync'd to Pub/Sub or any other services supported
+              by Cloud Logging.
+        """
+        super().__init__(user_emails=user_emails, enable_logging=enable_logging)

tests/system/aiplatform/test_model_monitoring.py

@@ -31,6 +31,7 @@
 
 # constants used for testing
 USER_EMAIL = "[email protected]"
+NOTIFICATION_CHANNEL = "projects/123/notificationChannels/456"
 PERMANENT_CHURN_MODEL_ID = "5295507484113371136"
 CHURN_MODEL_PATH = "gs://mco-mm/churn"
 DEFAULT_INPUT = {
@@ -90,10 +91,16 @@
 # global test constants
 sampling_strategy = model_monitoring.RandomSampleConfig(sample_rate=LOG_SAMPLE_RATE)
 
-alert_config = model_monitoring.EmailAlertConfig(
+email_alert_config = model_monitoring.EmailAlertConfig(
     user_emails=[USER_EMAIL], enable_logging=True
 )
 
+alert_config = model_monitoring.AlertConfig(
+    user_emails=[USER_EMAIL],
+    enable_logging=True,
+    notification_channels=[NOTIFICATION_CHANNEL],
+)
+
 schedule_config = model_monitoring.ScheduleConfig(monitor_interval=MONITOR_INTERVAL)
 
 skew_config = model_monitoring.SkewDetectionConfig(
@@ -149,7 +156,7 @@ def test_mdm_two_models_one_valid_config(self, shared_state):
             display_name=self._make_display_name(key=JOB_NAME),
             logging_sampling_strategy=sampling_strategy,
             schedule_config=schedule_config,
-            alert_config=alert_config,
+            alert_config=email_alert_config,
             objective_configs=objective_config,
             create_request_timeout=3600,
             project=e2e_base._PROJECT,
@@ -211,7 +218,7 @@ def test_mdm_pause_and_update_config(self, shared_state):
             display_name=self._make_display_name(key=JOB_NAME),
             logging_sampling_strategy=sampling_strategy,
             schedule_config=schedule_config,
-            alert_config=alert_config,
+            alert_config=email_alert_config,
             objective_configs=model_monitoring.ObjectiveConfig(
                 drift_detection_config=drift_config
             ),
@@ -284,7 +291,7 @@ def test_mdm_two_models_two_valid_configs(self, shared_state):
             display_name=self._make_display_name(key=JOB_NAME),
             logging_sampling_strategy=sampling_strategy,
             schedule_config=schedule_config,
-            alert_config=alert_config,
+            alert_config=email_alert_config,
             objective_configs=all_configs,
             create_request_timeout=3600,
             project=e2e_base._PROJECT,
@@ -338,7 +345,7 @@ def test_mdm_invalid_config_incorrect_model_id(self, shared_state):
                 display_name=self._make_display_name(key=JOB_NAME),
                 logging_sampling_strategy=sampling_strategy,
                 schedule_config=schedule_config,
-                alert_config=alert_config,
+                alert_config=email_alert_config,
                 objective_configs=objective_config,
                 create_request_timeout=3600,
                 project=e2e_base._PROJECT,
@@ -358,7 +365,7 @@ def test_mdm_invalid_config_xai(self, shared_state):
                 display_name=self._make_display_name(key=JOB_NAME),
                 logging_sampling_strategy=sampling_strategy,
                 schedule_config=schedule_config,
-                alert_config=alert_config,
+                alert_config=email_alert_config,
                 objective_configs=objective_config,
                 create_request_timeout=3600,
                 project=e2e_base._PROJECT,
@@ -388,7 +395,7 @@ def test_mdm_two_models_invalid_configs_xai(self, shared_state):
                 display_name=self._make_display_name(key=JOB_NAME),
                 logging_sampling_strategy=sampling_strategy,
                 schedule_config=schedule_config,
-                alert_config=alert_config,
+                alert_config=email_alert_config,
                 objective_configs=all_configs,
                 create_request_timeout=3600,
                 project=e2e_base._PROJECT,
@@ -399,3 +406,31 @@ def test_mdm_two_models_invalid_configs_xai(self, shared_state):
             "`explanation_config` should only be enabled if the model has `explanation_spec populated"
             in str(e.value)
         )
+
+    def test_mdm_notification_channel_alert_config(self, shared_state):
+        self.endpoint = shared_state["resources"][0]
+        aiplatform.init(project=e2e_base._PROJECT, location=e2e_base._LOCATION)
+        # test model monitoring configurations
+        job = aiplatform.ModelDeploymentMonitoringJob.create(
+            display_name=self._make_display_name(key=JOB_NAME),
+            logging_sampling_strategy=sampling_strategy,
+            schedule_config=schedule_config,
+            alert_config=alert_config,
+            objective_configs=objective_config,
+            create_request_timeout=3600,
+            project=e2e_base._PROJECT,
+            location=e2e_base._LOCATION,
+            endpoint=self.endpoint,
+        )
+
+        gapic_job = job._gca_resource
+        assert (
+            gapic_job.model_monitoring_alert_config.email_alert_config.user_emails
+            == [USER_EMAIL]
+        )
+        assert gapic_job.model_monitoring_alert_config.enable_logging
+        assert gapic_job.model_monitoring_alert_config.notification_channels == [
+            NOTIFICATION_CHANNEL
+        ]
+
+        job.delete()

tests/unit/aiplatform/test_model_monitoring.py

@@ -31,6 +31,7 @@
 _TEST_DRIFT_TRESHOLD = {"key": 0.2}
 _TEST_EMAIL1 = "test1"
 _TEST_EMAIL2 = "test2"
+_TEST_NOTIFICATION_CHANNEL = "projects/123/notificationChannels/456"
 _TEST_VALID_DATA_FORMATS = ["tf-record", "csv", "jsonl"]
 _TEST_SAMPLING_RATE = 0.8
 _TEST_MONITORING_INTERVAL = 1
@@ -105,10 +106,16 @@ def test_valid_configs(
             monitor_interval=_TEST_MONITORING_INTERVAL
         )
 
-        alert_config = model_monitoring.EmailAlertConfig(
+        email_alert_config = model_monitoring.EmailAlertConfig(
             user_emails=[_TEST_EMAIL1, _TEST_EMAIL2]
         )
 
+        alert_config = model_monitoring.AlertConfig(
+            user_emails=[_TEST_EMAIL1, _TEST_EMAIL2],
+            enable_logging=True,
+            notification_channels=[_TEST_NOTIFICATION_CHANNEL],
+        )
+
         prediction_drift_config = model_monitoring.DriftDetectionConfig(
             drift_thresholds=_TEST_DRIFT_TRESHOLD
         )
@@ -149,8 +156,17 @@ def test_valid_configs(
             == prediction_drift_config.as_proto()
         )
         assert objective_config.as_proto().explanation_config == xai_config.as_proto()
+        assert (
+            _TEST_EMAIL1 in email_alert_config.as_proto().email_alert_config.user_emails
+        )
+        assert (
+            _TEST_EMAIL2 in email_alert_config.as_proto().email_alert_config.user_emails
+        )
         assert _TEST_EMAIL1 in alert_config.as_proto().email_alert_config.user_emails
         assert _TEST_EMAIL2 in alert_config.as_proto().email_alert_config.user_emails
+        assert (
+            _TEST_NOTIFICATION_CHANNEL in alert_config.as_proto().notification_channels
+        )
         assert (
             random_sample_config.as_proto().random_sample_config.sample_rate
             == _TEST_SAMPLING_RATE