Data Selection and Filtering
The data selection and filtering components provide builders for operators that control which data is included in the fringe fitting process based on various selection criteria and quality metrics.
MHO_DataSelectionBuilder
Class |
|
Primary Functionality |
Builds a data selection operator |
Key Features |
Constructs and initializes data selection operator
Acts on fringe data for selection operations
Inherits from MHO_OperatorBuilder
|
The MHO_DataSelectionBuilder class builds a data selection operator that controls which data is included in the fringe fitting process. This operator can filter data based on AP-offset time windows, named frequency groups, specific channel labels, and/or polarization-product sets. Selection by minute-past-the-hour (positive start/stop values, per the legacy fourfit convention) is not yet supported; only negative start/stop values (seconds relative to scan start/stop) are currently honored, and positive values are rejected with an error and no selection is applied.
The builder provides flexible data selection capabilities for fringe fitting operations.
MHO_MinWeightBuilder
Class |
|
Primary Functionality |
Builds MHO_MinWeight operator for weight-based filtering |
Key Features |
Constructs and adds MHO_MinWeight operator to toolbox
Cuts data with weight less than threshold
Implements quality-based data filtering
|
The MHO_MinWeightBuilder class builds a minimum weight operator that filters data based on weight thresholds. This operator removes data points that have weights below a specified minimum value, effectively filtering out low-quality or unreliable data from the fringe fitting process. If the configured threshold is 0, the operator is not constructed at all (rather than applying a no-op threshold), so no weight-based filtering occurs.