Schedule Tracker
While the model tracker gives the accuracy levels of the model, the Schedule Tracker describes whether a training is successful or not each time the model is run, based on the data quality.
To review the schedule tracker, first navigate to the Schedule Tracker page as detailed below:
In the top left-hand corner of the main dashboard screen, click the Manage icon:
.On the Manage panel, click Recommendations
.Select Schedule Tracker.
The following information is displayed for each model.
Attribute | Description |
|---|---|
Domain | This is the functional domain that the model is associated with - for example, parking or lighting. |
Entity Type | This is the entity type that the model is associated with. |
Property | This is the parameter/observation for which the model is being trained. |
Purpose | This is the purpose of the model - predict or train. |
No of Entities | The total number of input data points. |
NAN Percentage | This indicates the percentage of missing or invalid data. If the NAN Percentage is high then the quality of the data will be low. |
Data Fetch Start | This is the time at which the data fetch started for that batch of training. |
Data Fetch End | This is the time at which the data fetch ended for that batch of training. |