Is your feature request related to a problem? Please describe
Initially, predictions and events dataframes are merged to align each row with the corresponding event row. Events outside the specified time window are then filtered out. However, enhancing traceability in this process would be beneficial, as it would allow for better tracking and understanding of how each target value (positive or negative case) is obtained.
E.g., if a negative case is obtained after filtering a detected event (for instance, scoring time is outside the time window) or there are no associated events found.
Describe the solution you'd like
Provide a way to distinguish between different "types" of negative cases in visualizations.
Is your feature request related to a problem? Please describe
Initially, predictions and events dataframes are merged to align each row with the corresponding event row. Events outside the specified time window are then filtered out. However, enhancing traceability in this process would be beneficial, as it would allow for better tracking and understanding of how each target value (positive or negative case) is obtained.
E.g., if a negative case is obtained after filtering a detected event (for instance, scoring time is outside the time window) or there are no associated events found.
Describe the solution you'd like
Provide a way to distinguish between different "types" of negative cases in visualizations.