Qa Network¶
Module: spectralbridge.qa.network.
Diagnostics for empirical sensor translations and translation networks.
These functions evaluate supplied predictions. They do not fit translation models and therefore cannot accidentally turn in-sample fits into validation evidence.
API glossary ยท Workflow / scientific context
Signatures and documentation below are collected from the current source. Advanced objects are not newly declared stable by inclusion here.
Audience: public. Also exported as spectralbridge.qa.translation_edge_metrics.
translation_edge_metrics ¶
translation_edge_metrics(observed: ndarray, predicted: ndarray) -> dict[str, float | int | None]
Evaluate one sensor-to-sensor edge from paired held-out observations.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
observed
|
ndarray
|
Paired measured/reference and predicted values in common units; nonfinite pairs are omitted and residuals are predicted minus observed. |
required |
predicted
|
ndarray
|
Paired measured/reference and predicted values in common units; nonfinite pairs are omitted and residuals are predicted minus observed. |
required |
Returns:
| Type | Description |
|---|---|
dict[str, float | int | None]
|
Paired residual metrics in the target units; high R-squared alone does not certify interchangeability. |
See Also
spectralbridge.qa.metrics.reflectance_summary : Related source-derived API contract.
Audience: public. Also exported as spectralbridge.qa.grouped_residual_metrics.
grouped_residual_metrics ¶
grouped_residual_metrics(observed: ndarray, predicted: ndarray, groups: Iterable[Any]) -> dict[str, Any]
Return residual metrics by an explicit validation block.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
observed
|
ndarray
|
Paired measured/reference and predicted values in common units; nonfinite pairs are omitted and residuals are predicted minus observed. |
required |
predicted
|
ndarray
|
Paired measured/reference and predicted values in common units; nonfinite pairs are omitted and residuals are predicted minus observed. |
required |
groups
|
Iterable[Any]
|
Aligned independent-unit identifiers; keep repeated observations or parent-connected mixtures in the same group. None uses the function-specific row-based convention described in Notes. |
required |
Returns:
| Type | Description |
|---|---|
dict[str, Any]
|
Explicit per-group residual summaries and an overall paired summary; the supplied group variable is not independently validated. |
Raises:
| Type | Description |
|---|---|
ValueError
|
If the supplied options, dimensions, support or provenance violate the input contract. |
Notes
groups should identify held-out sites, dates, campaigns, or flightlines.
The function reports, but does not invent, the blocking strategy.
See Also
spectralbridge.qa.metrics.reflectance_summary : Related source-derived API contract.
Audience: public. Also exported as spectralbridge.qa.path_consistency_metrics.
path_consistency_metrics ¶
path_consistency_metrics(direct_prediction: ndarray, indirect_prediction: ndarray, *, observed_target: ndarray | None = None) -> dict[str, Any]
Compare a direct translation with a multi-edge translation path.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
direct_prediction
|
ndarray
|
Matched target-sensor predictions from direct and multi-edge translation paths, in common units and row order. |
required |
indirect_prediction
|
ndarray
|
Matched target-sensor predictions from direct and multi-edge translation paths, in common units and row order. |
required |
observed_target
|
ndarray | None
|
Optional independently observed target values for residual summaries of each translation path. |
None
|
Returns:
| Type | Description |
|---|---|
dict[str, Any]
|
Direct-versus-indirect residual summary and optional each-path residuals against observed_target. |
See Also
spectralbridge.qa.metrics.reflectance_summary : Related source-derived API contract.
Audience: public. Also exported as spectralbridge.qa.cycle_consistency_metrics.
cycle_consistency_metrics ¶
cycle_consistency_metrics(starting_values: ndarray, cycled_values: ndarray) -> dict[str, float | int | None]
Measure drift after translating around a sensor cycle back to its start.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
starting_values
|
ndarray
|
Matched values before and after a translation cycle, expressed on the starting sensor support and scale. |
required |
cycled_values
|
ndarray
|
Matched values before and after a translation cycle, expressed on the starting sensor support and scale. |
required |
Returns:
| Type | Description |
|---|---|
dict[str, float | int | None]
|
Residual drift metrics comparing returned-cycle values with starting values in their common units. |
See Also
spectralbridge.qa.metrics.reflectance_summary : Related source-derived API contract.