Azure Machine Learning Service - Monitoring

Monitor Your Azure Machine Learning Experiments

Gain visibility into the performance of your Azure Machine Learning experiments. Track metrics, logs, and resource utilization to optimize your models and workflows.

Real-time Metrics

Track key metrics such as training time, GPU utilization, and memory consumption in real-time.

Log Management

Collect and analyze logs from your experiments to identify issues and improve model accuracy.

Examples

import logging
from azure.machinelearning.core import MetricCategory, MetricName, MetricValue, MetricUnit

# Define a metric category
metric_category = MetricCategory("TrainingTime")

# Define a metric name
metric_name = MetricName("TrainingDuration", "Seconds")

# Define a metric value
metric_value = MetricValue(60.5)

# Define a metric unit
metric_unit = MetricUnit.Seconds

# Create a MetricCategory object
$MetricCategory = New-AzMachineLearningJobMetricCategory -Name "TrainingTime"

# Create a Metric object
$Metric = New-AzMachineLearningJobMetric -JobName "MyExperiment" -MetricName "TrainingDuration" -MetricCategory $MetricCategory -Value 60.5 -Unit Seconds