Accessing the Machine Learning page


The ML page provides a centralized view of all Machine Learning (ML) models to identify, understand, and manage their machine learning models within the system.

To access the ML page

In BMC Helix Edge, navigate to Intelligence > Machine Learning.

The system displays the following page:

image-2025-1-27_12-40-59-1.png

The following table describes each model's configuration, training, and deployment status.

Column name

Description

Model name

The ML model name displays the name given to each ML model. The model name is the primary identifier of the model.

Description

The description Provides a brief description of the model's purpose. 

Algorithm

The ML algorithm type that is used in the model. The following algorithms are available:

  • AutoEncoder-Anomaly: Learns to reconstruct normal data, flagging anomalies as those poorly reconstructed.
  • RandomForest-Classification: Combines multiple decision trees to predict class labels.
  • Multivariate-DeepVar-Timeseries: Forecasts multiple time series variables using a deep learning model.
  • LightGBM-Regression: Efficient gradient boosting algorithm for accurate regression predictions.

Training status

The model's training status. The following statuses are available:

  • New: The model is newly added and has not been trained.
  • TrainingCompleted: The model has been trained.
  • Failed: The training process encountered an error.
  • TrainingInProgress: The ongoing training session on starting training the model.

Deployments

The model's deployment status, including the nodes where the model is deployed. The following deployment statuses  are available:

  • ModelDeployed: Indicates that BMC Helix Edge deployed a specific version of a Machine Learning model to a target environment.
  • ModelDeploymentFailed: The deployment process for a specific model version encountered an error. BMC Helix Edge failed to deploy the model to the target environment.
  • ReadyToDeploy: A model is eligible and prepared for deployment, but the actual deployment process is not yet initiated. The model has been trained or uploaded, and all the necessary configurations are in place.

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*