Monitoring Lookout for Vision with Amazon CloudWatch - Amazon Lookout for Vision

Monitoring Lookout for Vision with Amazon CloudWatch

You can use CloudWatch to monitor image processing and anomaly detecting metrics in Lookout for Vision.

Lookout for Vision metrics

Amazon Lookout for Vision sends the following metrics to CloudWatch for the DetectAnomalies action:

Metric Description

DetectedAnomalyCount

The number of anomalies detected in a project

Valid Statistics: Sum,Average

Unit: Count

ProcessedImageCount

The total number of images run through anomaly detection

Valid Statistics: Sum,Average

Unit: Count

InvalidImageCount

The number of images that were invalid and could not return a result

Valid Statistics: Sum,Average

Unit: Count

SuccessfulRequestCount

The number of successful API calls

Valid Statistics: Sum,Average

Unit: Count

ErrorCount

The number of API errors

Valid Statistics: Sum,Average

Unit: Count

ThrottledCount

The number of API errors that were due to throttling

Valid Statistics: Sum,Average

Unit: Count

Time

The time in milliseconds for Lookout for Vision to compute the anomaly detection

Valid Statistics: Data Samples,Average

Units: Milliseconds for Average statistics and Count for Data Samples statistics

MinInferenceUnits

The minimum number of inference units specified during the StartModel request.

Valid statistics: Average

Unit: Count

MaxInferenceUnits

The maximum number of inference units specified during the StartModelrequest.

Valid statistics: Average

Unit: Count

DesiredInferenceUnits

The number of inference units to which Lookout for Vision is scaling up or down.

Valid statistics: Average

Unit: Count

InServiceInferenceUnits

The number of inference units that the model is using.

Valid statistics: Average

It is recommended that you use the Average statistic to obtain the 1 minute average of how many instances are used.

Unit: Count

Dimensions of Lookout for Vision metrics

Lookout for Vision metrics use the Lookout for Vision namespace and provide metrics for the following dimensions:

Dimension Description

ProjectName

You can split metrics by project to see which projects are having problems or need to be updated.

ModelVersion

You can split metrics by model version to see which models are having problems or need to be updated.