Whenever a calculated feature is executed the observations of this calculated feature are modified. With instantiating calculated features this is different. Instantiating calculated features will create new features every time they are executed and will store the results of the calculation algorithm as observations of that newly created features. Instantiating calculated features are useful for creating snapshots of data at specific points in time.

The features generated are regular features of the same feature type as the calculated feature. The newly create features will be named according to following scheme:

<Instantiating calculated feature name> <YYYY-MM-DD HH:MM:SS +TimeOffset>

Domain and subtype of the newly created features are the same as in the calculated feature.

For each feature created from an instantiating calculated feature the application server creates a calculation relation.

Instantiating calculated features will only execute if the user who last edited the calculation algorithm has the permission to create features in the domain of the calculated feature.

If incremental execution is turned on for the calculation algorithm the newly created features will only include the results of the algorithm applied to the observations of source features that were added since the last execution.

Depending on the view structure it is possible that the newly created instance is not visible or not accessible, make sure that you have an appropriate view that will include the newly created features.