Selecting a specific branch of the ML model in the History section

The History section provides access to the history of incoming data, the results of data processing by Kaspersky MLAD, generated predictions, and registered incidents.

If the ML model used for a monitored asset has several elements for processing data, Kaspersky MLAD lets you select a specific branch of the ML model to display the operating results of the corresponding model element:

• For an ML model branch based on the Forecaster Detector, operating results are displayed as predicted values for specific tags, individual errors in the prediction of specific tags, and cumulative MSE and dot indicators of incidents registered by the detector.
• For an ML model branch based on the Rule Detector, operating results are presented as values obtained as a result of the work of diagnostic rules and dot indicators of incidents.
• An ML model branch is not created for the Limit Detector. The dot indicators of incidents registered using this detector are displayed if use of the Limit Detector is enabled and the display of indicators for all tags is enabled.

To display the predicted values of a tag on graphs in the History section, and to display the values obtained as a result of the work of diagnostic rules, you must customize the display of graphs.

To view the operating results of a specific ML model branch:

1. In the main menu, select the History section.
2. On the opened page, in the Model branch drop-down list, select the check boxes next to the relevant branches of the ML model.

   The names of the selected branches are displayed in the field.

   The branches belonging to the currently used ML model are located in the upper part of the list. The lower part of the list displays branches of other currently unused ML models that were uploaded to Kaspersky MLAD. An ML model branch is displayed in the drop-down list only after Kaspersky MLAD receives data that resulted from operations of the specific branch.

   The graphs of the selected preset will display the predicted values of tags or the values obtained as a result of the work of diagnostic rules depending on the type of detector in the selected ML model branch.

   If you need to hide the operating results from previously selected ML model branches, clear the check boxes next to these branches (however, one of the branches must remain active for graphs to be displayed in the History section).

3. If you need to display the MSE received as a result of data processing by a specific ML model branch:
   1. Click the settings button below the tag graphs on the left side of the page.
   2. In the MSE graph display settings pane that appears on the right, select the branch from the Model branch drop-down list. You can select only one ML model branch from the list.
   3. Click the Close button.

The MSE graph displays the MSE values for the selected branch of the ML model.

The lower part of the graph displays the dot indicators of incidents that were registered by the selected ML model branches. If the display of indicators for all tags is enabled, the dot indicators of incidents that were registered by all ML model branches will be displayed.

Page top