MetricsAdvisorClient Class

Reference

Represents an client that calls restful API of Azure Metrics Advisor service.

Inheritance: builtins.object

MetricsAdvisorClient

Constructor

MetricsAdvisorClient(endpoint: str, credential: MetricsAdvisorKeyCredential, **kwargs: Any)

Parameters

endpoint: str

Required

Supported Cognitive Services endpoints (protocol and hostname, for example: https://<resource-name>.cognitiveservices.azure.com).

credential: MetricsAdvisorKeyCredential or TokenCredential

Required

An instance of ~azure.ai.metricsadvisor.MetricsAdvisorKeyCredential. which requires both subscription key and API key. Or an object which can provide an access token for the vault, such as a credential from identity

Methods

add_feedback	Create a new metric feedback.
close	Close the MetricsAdvisorClient session.
get_feedback	Get a metric feedback by its id.
list_alerts	Query alerts under anomaly alert configuration.
list_anomalies	Query anomalies under a specific alert or detection configuration.
list_anomaly_dimension_values	Query dimension values of anomalies.
list_feedback	List feedback on the given metric.
list_incident_root_causes	Query root cause for incident.
list_incidents	Query incidents under a specific alert or detection configuration.
list_metric_dimension_values	List dimension from certain metric.
list_metric_enriched_series_data	Query series enriched by anomaly detection.
list_metric_enrichment_status	Query anomaly detection status.
list_metric_series_data	Get time series data from metric.
list_metric_series_definitions	List series (dimension combinations) from metric.

add_feedback

Create a new metric feedback.

add_feedback(feedback: Union[azure.ai.metricsadvisor.models._models.AnomalyFeedback, azure.ai.metricsadvisor.models._models.ChangePointFeedback, azure.ai.metricsadvisor.models._models.CommentFeedback, azure.ai.metricsadvisor.models._models.PeriodFeedback], **kwargs: Any) -> None

Parameters

feedback: AnomalyFeedback or ChangePointFeedback or CommentFeedback or PeriodFeedback

Required

metric feedback.

Exceptions

HttpResponseError

Examples

Add new feedback.


   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient
   from azure.ai.metricsadvisor.models import AnomalyFeedback, ChangePointFeedback, CommentFeedback, PeriodFeedback

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   anomaly_feedback = AnomalyFeedback(metric_id=metric_id,
                                      dimension_key={"Dim1": "Common Lime"},
                                      start_time=datetime.datetime(2020, 8, 5),
                                      end_time=datetime.datetime(2020, 8, 7),
                                      value="NotAnomaly")
   client.add_feedback(anomaly_feedback)

   change_point_feedback = ChangePointFeedback(metric_id=metric_id,
                                               dimension_key={"Dim1": "Common Lime"},
                                               start_time=datetime.datetime(2020, 8, 5),
                                               end_time=datetime.datetime(2020, 8, 7),
                                               value="NotChangePoint")
   client.add_feedback(change_point_feedback)

   comment_feedback = CommentFeedback(metric_id=metric_id,
                                      dimension_key={"Dim1": "Common Lime"},
                                      start_time=datetime.datetime(2020, 8, 5),
                                      end_time=datetime.datetime(2020, 8, 7),
                                      value="comment")
   client.add_feedback(comment_feedback)

   period_feedback = PeriodFeedback(metric_id=metric_id,
                                    dimension_key={"Dim1": "Common Lime"},
                                    start_time=datetime.datetime(2020, 8, 5),
                                    end_time=datetime.datetime(2020, 8, 7),
                                    period_type="AssignValue",
                                    value=2)
   client.add_feedback(period_feedback)

close

Close the MetricsAdvisorClient session.

close() -> None

Exceptions

HttpResponseError

get_feedback

Get a metric feedback by its id.

get_feedback(feedback_id: str, **kwargs: Any) -> Union[MetricFeedback, FeedbackUnion]

Parameters

feedback_id: str

Required

the id of the feedback.

Returns

The feedback object

Return type

MetricFeedback,

AnomalyFeedback,

ChangePointFeedback,

CommentFeedback,

PeriodFeedback

Exceptions

HttpResponseError

Examples

Get a metric feedback by its id.


   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   feedback_id = os.getenv("METRICS_ADVISOR_FEEDBACK_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   result = client.get_feedback(feedback_id=feedback_id)
   print("Type: {}; Id: {}".format(result.feedback_type, result.id))

list_alerts

Query alerts under anomaly alert configuration.

list_alerts(alert_configuration_id: str, start_time: datetime.datetime, end_time: datetime.datetime, time_mode: Union[str, AlertQueryTimeMode], **kwargs: Any) -> ItemPaged[AnomalyAlert]

Parameters

alert_configuration_id: str

Required

anomaly alert configuration unique id.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

Required

start time.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

Required

end time.

time_mode: str or <xref:azure.ai.metricsadvisor.models.AlertQueryTimeMode>

Required

time mode. Possible values include: "AnomalyTime", "CreatedTime", "ModifiedTime".

skip: int

Returns

AnomalyAlerts under anomaly alert configuration.

Return type

ItemPaged[AnomalyAlert]

Exceptions

HttpResponseError

Examples

Query anomaly detection results.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_alerts(
       alert_configuration_id=alert_config_id,
       start_time=datetime.datetime(2020, 1, 1),
       end_time=datetime.datetime(2020, 9, 9),
       time_mode="AnomalyTime",
   )

   tolist = []
   for result in results:
       tolist.append(result)
       print("Alert id: {}".format(result.id))
       print("Create time: {}".format(result.created_time))
   return tolist

list_anomalies

Query anomalies under a specific alert or detection configuration.

list_anomalies(alert_configuration_id, alert_id, **kwargs)

Parameters

alert_configuration_id: str

anomaly alert configuration unique id.

alert_id: str

alert id.

detection_configuration_id: str

anomaly detection configuration unique id.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

end time filter under chosen time mode.

skip: int

filter: DetectionAnomalyFilterCondition

Returns

Anomalies under a specific alert or detection configuration.

Return type

AsyncItemPaged[DataPointAnomaly]

Exceptions

HttpResponseError

list_anomaly_dimension_values

Query dimension values of anomalies.

list_anomaly_dimension_values(detection_configuration_id: str, dimension_name: str, start_time: Union[str, datetime.datetime], end_time: Union[str, datetime.datetime], **kwargs: Any) -> ItemPaged[str]

Parameters

detection_configuration_id: str

Required

anomaly detection configuration unique id.

dimension_name: str

Required

dimension to query.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

Required

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

Required

end time filter under chosen time mode.

skip: int

str] dimension_filter: <xref:Dict>[<xref:str,>

filter specfic dimension name and values.

Returns

Dimension values of anomalies.

Return type

ItemPaged[str]

Exceptions

HttpResponseError

Examples

Query dimension values.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   detection_configuration_id = os.getenv("METRICS_ADVISOR_DETECTION_CONFIGURATION_ID")
   dimension_name = "city"

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_anomaly_dimension_values(
       detection_configuration_id=detection_configuration_id,
       dimension_name=dimension_name,
       start_time=datetime.datetime(2020, 1, 1),
       end_time=datetime.datetime(2020, 10, 21),
   )
   for result in results:
       print(str(result))

list_feedback

List feedback on the given metric.

list_feedback(metric_id: str, **kwargs: Any) -> ItemPaged[Union[MetricFeedback, FeedbackUnion]]

Parameters

metric_id: str

Required

filter feedbacks by metric id

skip: int

dimension_key: dict[str, str]

filter specfic dimension name and values

feedback_type: str or FeedbackType

filter feedbacks by type. Possible values include: "Anomaly", "ChangePoint", "Period", "Comment".

datetime.datetime] start_time: <xref:Union>[<xref:str,>

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

end time filter under chosen time mode.

time_mode: str or FeedbackQueryTimeMode

time mode to filter feedback. Possible values include: "MetricTimestamp", "FeedbackCreatedTime".

Returns

Pageable list of MetricFeedback

Return type

ItemPaged[<xref:Union>[MetricFeedback, AnomalyFeedback, ChangePointFeedback, CommentFeedback, PeriodFeedback]]

Exceptions

HttpResponseError

Examples

List feedback on the given metric.


   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_feedback(metric_id=metric_id)
   for result in results:
       print("Type: {}; Id: {}".format(result.feedback_type, result.id))

list_incident_root_causes

Query root cause for incident.

list_incident_root_causes(detection_configuration_id: str, incident_id: str, **kwargs: Any) -> ItemPaged[IncidentRootCause]

Parameters

detection_configuration_id: str

Required

anomaly detection configuration unique id.

incident_id: str

Required

incident id.

Returns

Pageable of root cause for incident

Return type

ItemPaged[IncidentRootCause]

Exceptions

HttpResponseError

Examples

Query incident root causes.


   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   detection_configuration_id = os.getenv("METRICS_ADVISOR_DETECTION_CONFIGURATION_ID")
   incident_id = os.getenv("METRICS_ADVISOR_INCIDENT_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))
   results = client.list_incident_root_causes(
       detection_configuration_id=detection_configuration_id,
       incident_id=incident_id,
   )
   for result in results:
       print("Score: {}".format(result.score))
       print("Description: {}".format(result.description))

list_incidents

Query incidents under a specific alert or detection configuration.

list_incidents(alert_configuration_id, alert_id, **kwargs)

Parameters

alert_configuration_id: str

anomaly alerting configuration unique id.

alert_id: str

alert id.

detection_configuration_id: str

anomaly detection configuration unique id.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

end time filter under chosen time mode.

skip: int

filter: DetectionAnomalyFilterCondition

Returns

AnomalyIncidents under a specific alert or detection configuration.

Return type

AsyncItemPaged[AnomalyIncident]

Exceptions

HttpResponseError

list_metric_dimension_values

List dimension from certain metric.

list_metric_dimension_values(metric_id: str, dimension_name: str, **kwargs: Any) -> ItemPaged[str]

Parameters

metric_id: str

Required

metric unique id.

dimension_name: str

Required

the dimension name

skip: int

dimension_value_filter: str

dimension value to be filtered.

Returns

Dimension from certain metric.

Return type

ItemPaged[str]

Exceptions

HttpResponseError

Examples

Query metric dimension values.


   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")
   dimension_name = "city"

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_metric_dimension_values(
       metric_id=metric_id,
       dimension_name=dimension_name,
   )
   for result in results:
       print(str(result))

list_metric_enriched_series_data

Query series enriched by anomaly detection.

list_metric_enriched_series_data(detection_configuration_id: str, series: Union[List[SeriesIdentity], List[Dict[str, str]]], start_time: Union[str, datetime.datetime], end_time: Union[str, datetime.datetime], **kwargs: Any) -> ItemPaged[MetricEnrichedSeriesData]

Parameters

detection_configuration_id: str

Required

anomaly alerting configuration unique id.

series: SeriesIdentity or list[dict[str, str]]

Required

List of dimensions specified for series.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

Required

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

Required

end time filter under chosen time mode.

Returns

Pageable of MetricEnrichedSeriesData

Return type

ItemPaged[MetricEnrichedSeriesData]

Exceptions

HttpResponseError

Examples

Query metric enriched series data.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   detection_configuration_id = os.getenv("METRICS_ADVISOR_DETECTION_CONFIGURATION_ID")
   series_identity = {"city": "Los Angeles"}

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_metric_enriched_series_data(
       detection_configuration_id=detection_configuration_id,
       start_time=datetime.datetime(2020, 9, 1),
       end_time=datetime.datetime(2020, 11, 1),
       series=[series_identity]
   )
   for result in results:
       print(str(result))

list_metric_enrichment_status

Query anomaly detection status.

list_metric_enrichment_status(metric_id: str, start_time: Union[str, datetime.datetime], end_time: Union[str, datetime.datetime], **kwargs: Any) -> ItemPaged[EnrichmentStatus]

Parameters

metric_id: str

Required

filter feedbacks by metric id.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

Required

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

Required

end time filter under chosen time mode.

skip: int

Returns

Anomaly detection status.

Return type

ItemPaged[EnrichmentStatus]

Exceptions

HttpResponseError

Examples

Query metric enrichment status.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_metric_enrichment_status(
           metric_id=metric_id,
           start_time=datetime.datetime(2020, 1, 1),
           end_time=datetime.datetime(2020, 10, 21),
       )
   for result in results:
       print(str(result))

list_metric_series_data

Get time series data from metric.

list_metric_series_data(metric_id: str, series_keys: List[Dict[str, str]], start_time: Union[str, datetime.datetime], end_time: Union[str, datetime.datetime], **kwargs: Any) -> ItemPaged[MetricSeriesData]

Parameters

metric_id: str

Required

metric unique id.

series_keys: list[dict[str, str]]

Required

query specific series.

datetime.datetime] start_time: <xref:Union>[<xref:str,>

Required

start time filter under chosen time mode.

datetime.datetime] end_time: <xref:Union>[<xref:str,>

Required

end time filter under chosen time mode.

Returns

Time series data from metric.

Return type

ItemPaged[MetricSeriesData]

Exceptions

HttpResponseError

Examples

Query metrics series data.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_metric_series_data(
           metric_id=metric_id,
           start_time=datetime.datetime(2020, 1, 1),
           end_time=datetime.datetime(2020, 10, 21),
           series_keys=[
               {"city": "Los Angeles", "category": "Homemade"}
           ]
       )
   for result in results:
       print(str(result))

list_metric_series_definitions

List series (dimension combinations) from metric.

list_metric_series_definitions(metric_id: str, active_since: datetime.datetime, **kwargs: Any) -> ItemPaged[MetricSeriesDefinition]

Parameters

metric_id: str

Required

metric unique id.

active_since: datetime

Required

Required. query series ingested after this time, the format should be yyyy-MM-ddTHH:mm:ssZ.

skip: int

dimension_filter: dict[str, list[str]]

filter specfic dimension name and values.

Returns

Series (dimension combinations) from metric.

Return type

ItemPaged[<xref:azure.ai.metricsadvisor.models.MetricSeriesDefinition>]

Exceptions

HttpResponseError

Examples

Query metric series definitions.


   import datetime
   from azure.ai.metricsadvisor import MetricsAdvisorKeyCredential, MetricsAdvisorClient

   service_endpoint = os.getenv("METRICS_ADVISOR_ENDPOINT")
   subscription_key = os.getenv("METRICS_ADVISOR_SUBSCRIPTION_KEY")
   api_key = os.getenv("METRICS_ADVISOR_API_KEY")
   metric_id = os.getenv("METRICS_ADVISOR_METRIC_ID")

   client = MetricsAdvisorClient(service_endpoint,
                                 MetricsAdvisorKeyCredential(subscription_key, api_key))

   results = client.list_metric_series_definitions(
           metric_id=metric_id,
           active_since=datetime.datetime(2020, 1, 1),
       )
   for result in results:
       print(str(result))