Metric class

Definition

Metric object for telemetry usage.

Use metrics for collecting and aggregating data that can be best aggregated into buckets for analysis.

Metric(name: str, value: float, required_fields: azureml.telemetry.contracts._required_fields.RequiredFields, standard_fields: azureml.telemetry.contracts._standard_fields.StandardFields, extension_fields: typing.Union[azureml.telemetry.contracts._extension_fields.ExtensionFields, NoneType] = None, metric_type: typing.Union[applicationinsights.channel.contracts.DataPointType.DataPointType, NoneType] = None, count: typing.Union[int, NoneType] = None, metric_min: typing.Union[float, NoneType] = None, metric_max: typing.Union[float, NoneType] = None, std_dev: typing.Union[float, NoneType] = None, *args: typing.Any, **kwargs: typing.Any)
Inheritance
builtins.dict
azureml.telemetry.contracts._telemetry_object_base.TelemetryObjectBase
Metric

Parameters

name
str

The name of the metric captured.

value
float

The value of the metric captured.

required_fields
RequiredFields

Required fields for the schema.

standard_fields
StandardFields

Standard fields for the schema.

extension_fields
ExtensionFields

Extension fields a.k.a Part C.

metric_type
applicationinsights.channel.contracts.DataPointType

The type of the metric.

count
int

The number of metrics that were aggregated into this data point.

metric_min
float

The minimum of all metrics collected that were aggregated into this data point.

metric_max
float

The maximum of all metrics collected that were aggregated into this data point.

std_dev
float

The standard deviation of all metrics collected that were aggregated into this data point.