Posted On: Dec 20, 2022
Amazon Lookout for Equipment analyzes equipment sensor data to train and build a machine learning model for your equipment – with no ML expertise required. Lookout for Equipment uses your unique ML model(s) and in real-time helps accurately identify early warning signs that could lead to machine failures. This helps you detect equipment abnormalities with speed and precision, quickly diagnose issues, and take action to reduce expensive downtime.
We are excited to announce we now allow customers to provide feedback about events detected by Lookout for Equipment via labels and label groups. Developers, through the API, can create, delete, describe, list and update labels and label groups associated with specific events. This allows end users, through labels and label groups, to provide a list of acceptable fault codes, start and end time of event(s) of interest, target equipment, ratings, and notes.
Label groups can then be used directly and programmatically when training a model instead of providing a CSV file located on S3. The information provided by these labels will still be used as periods of time where models will discard the data as abnormal and won’t train against it.
To learn more about this capability, see our API operations documentation. This new set of capabilities is available in all Regions where Amazon Lookout for Equipment is publicly available. For more information about region availability, see AWS Regional Services.